1. business and industrial
2. energy
3. natural gas

Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology 1301
Exposing the Dark Microbial
Biosphere
VALERIE HUBALEK
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2015
ISSN 1651-6214
ISBN 978-91-554-9363-9
urn:nbn:se:uu:diva-264115
Dissertation presented at Uppsala University to be publicly examined in Friessalen,
Evolutionary Biology Centre (EBC), Norbyvägen 14, Uppsala, Friday, 20 November 2015 at
09:00 for the degree of Doctor of Philosophy. The examination will be conducted in English.
Faculty examiner: Prof. Tim Urich (University of Greifswald, Institute for Microbiology).
Abstract
Hubalek, V. 2015. Exposing the Dark Microbial Biosphere. Digital Comprehensive
Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1301. 45 pp.
Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-554-9363-9.
Dark biosphere research has been widely neglected, although by volume this biome comprises
the lion’s share of habitats on our planet. In these systems the main metabolic strategies are of
chemotrophic nature, leading to gradual depletion of redox gradients essential for sustaining
life. Thus these environments are regarded more or less close to chemical equilibrium.
Here, we use sequence data of whole community metagenomes and taxonomic marker
approaches to study the ecology of environments close to the thermodynamic limit: deep
terrestrial aquifers and aphotic systems impacted by petroleum- derived products. We show
that these systems select for individuals with reduced genomes and cell sizes, likely as a mode
to save energy. Due to genome reduction, these so called “streamlined” cells are reduced
in the number of genes and metabolic pathways. This loss has led to community members
sharing the metabolic burden of synthesizing in particular energy costly metabolites, creating
tight interdependencies between the community members, as a consequence. In addition, we
propose that cells scavenging anabolic products derived from detrital biomass and intermediate
fermentation products are equally important in these systems. Hence, life at the thermodynamic
limit involves a much more complex biological system than previously shown, that goes beyond
traditionally described electron- and intermediate metabolite-transfer dependencies.
This thesis furthermore includes ecological implications, demonstrating how species diversity
and community metabolism are shaped by redox gradients and dispersal potential in the deep
biosphere and contaminated sediments. This research is also relevant from a practical point
of view, as it pinpoints new opportunities for enhanced bioremediation through metabolite
additions in order to raise the efficiency of degradation processes.
Valerie Hubalek, Department of Ecology and Genetics, Norbyvägen 18 D, Uppsala
University, SE-752 36 Uppsala, Sweden.
© Valerie Hubalek 2015
ISSN 1651-6214
ISBN 978-91-554-9363-9
urn:nbn:se:uu:diva-264115 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-264115)
List of Papers
This thesis is based on the following papers, which are referred to in the text
by their Roman numerals.
I
Xiaofen Wu, Karin Holmfeldt, Valerie Hubalek, Daniel Lundin,
Mats Åström, Stefan Bertilsson and Mark Dopson. The microbial
interactome of the terrestrial deep biosphere reveals metabolic
partitioning among populations. ISME Journal, In Press
II
Valerie Hubalek, Xiaofen Wu, Alexander Eiler, Moritz Buck, Christine Heim, Mark Dopson, Stefan Bertilsson and Danny Ionescu.
Connectivity driven bacterial diversity patterns and functional
potential in three deep aquifers of the Fennoscandian shield.
Submitted.
III Valerie Hubalek & Moritz Buck, BoonFei Tan, Julia Foght, Annelie
Wendeberg, David Berry, Stefan Bertilsson and Alexander Eiler.
Metabolic partitioning in an alkane degrading bioreactor operating under methanogenic condition. Manuscript.
IV Valerie Hubalek & Inga Richert, Sari Peura, Jimmy Saw, Annelie
Wendeberg, Thijs Ettema and Stefan Bertilsson. Functional potential of microbial communities in freshwaer sediments affected by
century-long tar-contamination. Manuscript.
“Because the history of evolution is that life escapes all barriers. Life breaks
free. Life expands to new territories. Painfully, perhaps even dangerously.
But life finds a way."
Michael Crichton - Jurassic Park (1990)
Dedicated to Darkness
Cover designed by Harold Khan
List of additional Papers and Manuscripts
I
Karolin Tischer, Martha Schattenhofer, Ingo Fetzer, Valerie Hubalek, Hauke Harms and Annelie Wendeberg. Spatial distribution
of fermenting bacteria and methanogens involved in petroleum
hydrocarbon degradation in a BTEX contaminated aquifer.
Submitted.
II
Martha Schattenhofer, Valerie Hubalek, Annelie Wendeberg. (2014)
Identification of Microorganisms in Hydrocarbon-Contaminated
Aquifer Samples by Fluorescence In Situ Hybridization (CARDFISH). Hydrocarbon and Lipid Microbiology Protocols, Springer
Protocols Handbooks, DOI 10.1007/8623_2014_9
Contents
1 Introduction .............................................................................................. 11 1.1 The Unseen World ............................................................................ 11 1.2 The Dark Biosphere .......................................................................... 11 1.2.1 Terrestrial Deep Aquifers ........................................................... 12 1.2.2 Petroleum Environments ........................................................... 13 1.3 Microbial Metabolism in the Darkness ............................................ 15 1.3.1 Anaerobic metabolic cooperation: Syntrophy ............................ 15 1.3.2 Petroleum hydrocarbons as food and energy source................... 16 1.3.3 Energy- and nutrient cycling in the deep biosphere .................... 17 1.3.4 The start- and end products of the carbon cycle: Methane and
Carbon Dioxide .................................................................................... 19 1.4 Metagenomics and the Uncultured World .......................................... 20 2 The Present Thesis ................................................................................... 22 2.1 Aims.................................................................................................... 22 2.2 Methods .............................................................................................. 23 2.2.1 Study Sites .................................................................................. 23 2.2.2 Analytical Methods ..................................................................... 25 2.3 Results ................................................................................................ 26 2.3.1 Exploring the Deep Terrestrial Biosphere .................................. 26 2.3.2 Studies on Petroleum impacted environments ............................ 30 2.4 Conclusions and Outlook.................................................................... 33 3 Summary in Swedish ................................................................................ 36 4 Acknowledgements .................................................................................. 40 5 References ................................................................................................ 42 Abbreviations
16Sr RNA
Äspö HRL
ASS
BLAST
BSS
BTEX
CH4
CO2
DNA
Fe
H2ase
HSH2SO4
MAS
Mn
NGS
N2
NH4+
NMS
NO3NO2PAH
SLiMEs
SO3-
16S Ribosomal Ribonucleic Acid
Äspö Hard Rock Laboratory
Alkylsuccinate synthase
Basic Local Alignment Tool
Benzylsuccinate Synthase
Benzene, Toluene, Ethylbenzene and Xylene
Methane
Carbon dioxide
Deoxyribonucleic Acid
Iron
Hydrogenase
Hydrogensulfide
Sulfate
Methyl-Alkylsuccinate Synthase
Manganese
Next Generation Sequencing
Nitrogen gas
Ammonium
Napthyl-methylsuccinate synthase
Nitrate
Nitrite
Polycyclic aromatic hydrocarbons
Subsurface Lithoautotrophic Microbial Ecosystems
Sulfite
List of Figures
Figure 1. The redox tower of energy gradients
Figure 2. Alkane degradation and recycling of propionyl-CoA to fumarate
Figure 3. The geo-gas driven deep biosphere
Figure 4. The Äspö Hard Rock Laboratory (HRL) sampling site
Figure 5. The Syncrude Mildred Lake plant
Figure 6. Model of potential metabolic pathways in deep aquifers
Figure 7. Alpha diversity measures in deep aquifers
Figure 8. Model of a syntrophic network and associated secondary partners
Figure 9. The microbial community composition in Lake Grötingen
1 Introduction
1.1 The Unseen World
Microorganisms represent a central component of our planet. Historically
they were the first life forms to emerge and quantitatively they are dominant
by populating every single habitat on the planet, starting in the gut of the
tiniest insect1 (or on the in- and outside of other organisms) and deep sea
vents with boiling temperatures, to oil contaminated environments and rocks
many kilometers beneath the surface of the Earth.
They are central to the food web and mediate processes ranging from photosynthesis to decomposition of organic and inorganic matter, and play central roles in all major elemental cycles. Needless to say, all life on Earth
quintessentially depends on these microscopic entities as primary source of
nutrients, recyclers of dead matter and by their ability to metabolize even
toxic compounds2.
Studying the microbial world does not only offer us endless potential for
advancing technology in the medical or industrial area, but also gives us
insight into the beginning of life and its limitations. Limits that have been
and will likely also continue to be redefined.
1.2 The Dark Biosphere
The dark biosphere constitutes the largest collection of habitats by volume
on our planet, mainly comprising the aphotic zones of all major water bodies, the oceanic sub-seafloor sediments and the continental subsurface.
Although being such a big fraction of the biosphere, the lion’s share of
microbial studies conducted thus far has been directed towards light-exposed
environments. The boundaries for the truly dark environment which is supposedly independent of sunlight driven production, are difficult to define as
the products of photosynthesis are pervasive and will likely reach, at least to
some degree, most biotic habitats on our planet. Perhaps the only clear distinction between the photosphere and the dark biosphere is that the prevailing metabolic processes in dark systems are of chemotrophic nature and
more or less moving towards the thermodynamic limit. Underlying chemical
reactions that sustain life in these environments, are restricted by low substrate concentration, the availability of catalysts, as well as available reduct11
ants and oxidants yielding a relatively low free Gibb’s energy. Thus we consider these systems to move towards the thermodynamic and kinetic limits
for sustaining life.
1.2.1 Terrestrial Deep Aquifers
Major technical advances in the last few years have enabled a more extensive exploration of the terrestrial deep subsurface. In contrast to the first
studies of sub-seafloor microbiology, this research was initially facilitated by
the quest of finding oil or natural gas, minerals and water depositories3. Another reason was the design and validation of safe long-term storage facilities for nuclear waste as for the Äspö Hard Rock Laboratory (HRL) in
Southern Sweden3 (more details in following chapter). The aspect of studying resource and contaminant degradation as well as waste container stability
from a microbiological point of view were the next major motivation for the
initiation of terrestrial deep biosphere research3.
The need for drilling or tunnel excavation to reach these environments
pose an obvious risk for introducing contamination by adding potential electron-acceptors from the drilling equipment and drilling fluids and by altering
the salinity or other chemical components by draining the aquifer4. Various
strategies to minimize these effects are being applied, for example by adding
tracer substances to the drill fluids in order to test if contaminating drill water has intruded the aquifer4. Triple tube drilling can to some extent protect
aquifer surfaces from the drill water, but the risk for washout of fracture
materials and contamination still remains. Without the availability of undisturbed controls, the significance of these potential effects are difficult to
evaluate5.
However, the finding of 3.5 billion-year-old fossilized biofilm at 450 m
depth at the Äspö HRL provides encouraging evidence that microorganisms
found in granitic aquifers are not merely contamination artifacts6. Furthermore, this could suggest that hard rock aquifers might be one of the oldest
habitats for microbial life on the planet. It is plausible that before the atmosphere of the Earth was formed, the subsurface of our early planet sheltered
the first life forms from high levels of radiation7. This concept inspired
NASA’s scientists to look for life on other planets in deep underground settings in their future explorations3.
The definition of what is regarded as “deep” and “shallow” is most rationally
defined based on the degree of connectivity to the surface rather than actual
depth8.
The chemical properties of deep aquifers are determined by the origin and
mixing of the groundwater, which is essentially the result of the surrounding
geology. Groundwater-conducive fractures can range in size from micrometers to crush zones in faults that can hold large quantities of water. Addition12
ally, these systems can vary in the degree of isolation, not only from the
surface, but also from other aquifers that can replenish electron-donors and
more favorable electron acceptors5.
Undoubtedly, the abundance and activity of microbes in the deep subsurface are orders of magnitude lower than in surface sediments, as the energy
yield of photosynthetic derived processes when compared to chemosynthesis
is higher and thus produces more biomass at a faster rate. Despite the slow
growth, microbial biomass from the subseafloor has been estimated to
2.9x1029 cells, corresponding to 4.1 petagram (Pg) of carbon, which is
~0.6% of Earth’s total living biomass9.
It has been proposed that microorganisms residing at these great depths
use hydrogen and methane of primarily abiotic and to a lesser extent biogenic origin as energy sources to reduce ubiquitous carbon dioxide for autotrophic growth. Several studies have documented the general presence of
hydrogen and methane in the Fennoscandian Shield, strengthening the hypothesis of a “geo-gas” driven deep biosphere10–12. These microorganisms
have been termed subsurface lithoautotrophic microbial ecosystems
(SLiMEs) in the 1990s13.
A very recent finding and likely another important factor shaping the
SliMEs, which is also a strong indicator for their activity and growth, is the
relatively high abundance and diversity of viruses, which were found in the
Äspö HRL groundwater14. These viruses, aside from controlling the numbers
and activity of their prey, might cause a release of biomolecules from induced lysis of cells. This might provide an essential and highly available
carbon and nutrient source for the rest of the community. Moreover, phages
are known to be important vehicles for gene transfer in other ecosystems and
may well play a similar role in the deep biosphere14.
1.2.2 Petroleum Environments
Petroleum is organic matter (biomass) that once got buried and embedded in
the Earth’s crust. These organic carbon pools are thus also part of the deep
biosphere. Protected from the biotic carbon cycle of the surface, detritus was
subsequently transformed by long-term geological processes to its present
form15. This diverse mixture of compounds may reach the surface as a result
of geological events (e.g. earthquakes) or anthropogenic activities (e.g. drilling). Accordingly they are among the world’s most problematic and widespread contaminants. Global estimates suggest that about 47% of the crude
oil that enters the marine environment is from natural seeps, whereas 53%
results from leaks and spills during the extraction, transportation, refining,
storage, and utilization of petroleum16. Given the impact that recent oil spill
accidents have on the environment combined with their carcinogenic and
neurotoxic potential for humans, research on the distribution, degradation
and persistence of these compounds is of outmost importance.
13
Petroleum is one of the most complex mixtures of organic compounds
with more than 17.000 distinct chemical components. Crude oils feature
unique and variable chemical and physical characteristics, making it difficult
to predict their specific environmental fate. In broad terms, there are four
classes of constituents: the saturates, the aromatics, the asphaltenes (phenols,
fatty acids, ketones, esters, and porphyrins), and the resins (pyridines, quinolines, carbazoles, sulfoxides, and amides)17. Resin and asphaltene fractions
contain trace amounts of nitrogen, sulfur and/or oxygen in addition to the
bulk elemens carbon and hydrogen. These compounds in turn often form
complexes with heavy metals17.
The good news is that the highly reduced hydrocarbons in petroleum represent a rich source of energy for some microbes. Therefore, petroleum constituents are biodegradable when suitable electron acceptors are present.
Nevertheless, and this is the bad news, aromatic hydrocarbons and polar
fractions, which are very toxic, represent the most persistent fraction because
of high molecular stability and physical properties making them less susceptible to enzyme attack. This can lead to decade- or century-long residence
times in the environment with far-reaching consequences for human health
and the environment18.
Monoaromatic hydrocarbons referred to as BTEX (benzene, toluene,
ethylbenzene and xylene) are highly volatile and soluble substances commonly found in gasoline19. Polycyclic aromatic hydrocarbons (PAHs) are
compounds built on fused benzene rings and the most difficult to biodegrade. The range of environmental significance for biodegradation lies between naphthalene (C10H8) and coronene (C24H12). Anthropogenic sources
include combustion of fossil fuels and waste or other industrial activities and
is known to significantly affect coastal and inland surface waters as a result
of their ability to be transported over long distances as gases or aerosols20.
The first steps of hydrocarbon biodegradation result in the loss of the saturated compounds, while aromatic hydrocarbons and the polar fractions are
more resistant to biodegradation. Saturated hydrocarbons constitute the largest fraction of crude oil by mass, thus the biodegradation of saturated hydrocarbons is quantitatively the most important process in the removal of crude
oil from the environment. Overall, the susceptibility of hydrocarbons to microbial degradation is ranked as follows: linear alkanes > branched alkanes >
small aromatics > cyclic alkanes21.
Aside from hydrocarbons, petroleum reservoirs contain a wide range of
other organic substrates, that can serve as carbon and energy sources for
microorganisms. They include fatty acids and especially its main synthesis
and degradation intermediate acetate, formate, propionate and butyrate, and
more complex organic acids such as naphtenic acids with concentrations in
crude oil up to 100 mM21. All these substrates may be oxidized under anaerobiosis in the presence of terminal electron acceptors possibly present in
oilfield waters (e.g. sulfate).
14
1.3 Microbial Metabolism in the Darkness
The base of any food web starts with primary producers, autotrophs ("selffeeding", from the Greek autos "self" and trophe "nourishing") which per
definition can either use light or inorganic substances as energy sources and
a reductant (usually water or just hydrogen) to reduce CO2 from the surroundings in order to build up complex carbohydrates (such as glucose).
These compounds are then being further oxidized to release energy for synthesis of proteins, lipids and nucleic acids that also derive from primary production. Primary producers are then being consumed by heterotrophs, subsequently distributing these organic molecules across the food web.
Organisms that are cut-off from both photons and oxygen, such as organisms residing in the bottom layers of open water systems and deep aquifers,
will have to rely on electron acceptors with lower thermodynamic yield than
oxygen. In a sequential order, according to the free energy yielded by their
reduction, that will be NO3- which can be reduced to N2 or NH4+, Mn(IV) to
Mn(II), Fe(III) to Fe(II), SO42+ to HS- or S2- and at the end of the spectrum
CO2 being reduced to CH4 (Fig. 1)22. Thus, in terms of reactions, these are
ordered in the following manner: Denitrification, Mn(IV) or Fe(III) reduction, sulfate reduction and finally methanogenesis. The further down in the
redox cascade, the smaller the energy yield of the respective reaction, which
typically results in a slower growth and low cell division rates.
Figure 1. The redox tower showing the spectrum of energy gradients. EA= Electron
acceptor, EY= energy yield, RS= reduced species form of the electron acceptor.
1.3.1 Anaerobic metabolic cooperation: Syntrophy
Syntrophy is regarded as a version of symbiosis that is beneficial for one or
both of the interacting organisms, but is restricted to metabolic interactions23.
In the context of anoxic carbon cycling, the central syntrophic relationship consists of methanogens and fermenting organisms. The syntrophic
fermentor provides actetate, hydrogen and carbon dioxide, while the meth15
anogenic partner is responsible for efficient removal of hydrogen and formate, which would otherwise hinder fermentation because of thermodynamic constraints24. As methane production is energetically the least favorable
process, it requires efficient metabolic coordination with its partners. Many
syntrophs are known to rely on the capacity to perform reverse electron
transport-driven energy-conserving H2 production through electronconfurcating hydrogenase (H2ase) in combination with reduced ferredoxin25.
1.3.2 Petroleum hydrocarbons as food and energy source
Petroleum hydrocarbons, being highly reduced organic molecules, can be a
good substrate for microbial growth under the right conditions. With oxygen
being available, most hydrocarbons are susceptible to enzymatic attack using
monooxygenases (for alipathic and some aromatic hydrocarbons) and dioxygenases (for aromatic hydrocarbons), in processes where O2 acts both as an
electron acceptor and co-substrate26.
Under anaerobic conditions, the breakdown of hydrocarbons can be accomplished by anaerobic respiration with nitrate, ferric iron and sulfate. Under methanogenic conditions, hydrocarbon degradation is thermodynamically feasible only in syntrophic association with hydrogen-consuming microorganisms, such as methanogens or sulfate reducers27. To make the compound chemically reactive, several activation reactions have been proposed:
carboxylation28, hydroxylation29 and radicalization by fumarate addition. The
latter is the only proven mechanism so far, in which a family of glycyl radical enzymes are involved in the activation of structurally different hydrocarbons30: Alkylsuccinate synthase (ASS) mainly found under sulfate- and nitrate-reducing conditions31, benzylsuccinate synthase (BSS) reported under
methanogenic, sulfate-, nitrate- and iron-reducing conditions32 and methylalkylsuccinate synthase (MAS) reported under nitrate-reducing conditions33,
in addition to the non-homolog naphthyl-methylsuccinate synthase (NMS).
The primary substrates of ASS are n-alkanes, while BSS binds to toluene
and NMS on 2-methylnaphthalene. These enzymes add the respective substrate to the double bond of a fumarate molecule, forming a succinylated
product. The following reduction to acetyl-CoA is analogous for aromatic
hydrocarbons after ring-cleavage (Fig. 2). After alkane activation with
fumarate, the carbon skeleton is rearranged in order to allow CO2 demerging
via decarboxylation during which a free coenzyme A is being attached (likely produced by methanogens). After beta-oxidation the resulting acetyl-CoA
is demerged from the chain resulting in the removal of two carbons. Betaoxidation is then repeated until the endproducts are one acetyl-CoA or propionyl-CoA. The recycling of propionyl-CoA to fumarate is further demonstrated in Fig. 2.
16
Acetyl-CoA can further feed into acetotrophic methanogenesis, and the
hydrogen that is produced in parallel, can be incorporated by hydrogenotrophic methanogens (see section 1.3.4).
>?@'
!"
>?D@'
:6;4+453
:6;4+453"
/01'"!23"2%&'
'''''''''
#$%&'"
(=!!8&",*'4*:64#%-*&"(*'
'#$%"
>?C@'
><="
>A@'
(=!!8&6+0/%9'(6&,:*,"(*'
>%=!"=-"
!"#$%&'()*+*,%&'
#*"##"&-*.*&,'
''''''''
>?B@'
>B@'
.*,:6+."+%&6+0/%9'.=,"(*'
.*,:6+."+%&6+0/%9'.=,"(*
4*!"#$%56+"2%&'
''
>?A@'
>C@'
#$%"
<#%<8%&6+0/%9'!"#$%56+"(*'
"!6+0/%9'4*:64#%-*&"(*'
#$%&'"
$<="
2('"
70'%584"2%&'
>D@'
$%=!"=-"
"
>??@'
>?I@'
&'("
!"
*&%6+0/%9':64#","(*'
'"94+?,."@A,+53.38"49/012,$"
>H@'
>E@'
*+,*-,./012,$"
"!6+0/%9'"!*,6+0,#"&(;*#"(*'
70'%584"2%&'
+3*345"6.70"3.8"*+,8695"4935/012,$"
,+"*+,*-,./012,$"
>F@'
)$%!"
)$%&"
70:64#%56"!6+0/%9'4*:64#%-*&"(*'
"!*,%,#%<:8!'.*,:"&%-*&*(8('
>G@'
4935/012,$"
Figure 2. Alkane activation and formation to acetyl-CoA or acyl-CoA (steps 1-9)
and subsequent recycling of propionyl-CoA to fumarate (11-15). Red indicating
newly formed bonds, grey signifying dissociating molecules and coloured boxes are
for products that feed into other reactions or pathways.
1.3.3 Energy- and nutrient cycling in the deep biosphere
The deep subsurface is mainly anaerobic and the supply of organic carbon is
scarce. Hence it has been proposed that “geo-gases” of abiotic origin will be
of higher importance34. Highly abundant hydrogen and methane would in
this case function as electron donors and carbon dioxide as an oxidizing
complement. Part of the methane can be of biogenic origin, whereas H2 can
be formed abiotically mainly through radiolysis of water34 and to a minor
extent through rock transformation11. Migration of methane through faults
and fracture zones in the Fennoscandian Shield to the surface has been reported11, showing that there can be a flow of gases from the deep that travels
17
upwards. This provides strong evidence that the electron donors H2 and CH4
are broadly present in the deep subsurface.
Thus, with water available and under moderate temperature conditions,
microbes can in theory survive independent of light-driven processes (Fig.
3)34. Methanogens would use H2 to reduce CO2 to CH4, and acetogenic bacteria would also require these two components to produce acetate. This acetate can then be metabolized by acetoclastic methanogens, iron and sulfate
reducing bacteria and heterotrophs. This would position them as primary
producers in these systems34.
Figure 3. The geo-gas driven deep biosphere. Modified after Karsten Pedersen et
34
al.
Under certain circumstances, it has been hypothesized that this process can
be reversed. If a shortage in hydrogen is being created, for example when
methanogens co-occur with more fast-growing chemoautotrophs, such as
sulfate reducers, the opposite reaction would become more favorable and
methane or acetate would be stepwise oxidized anaerobically to CO235.
Not only energy and carbon are required for microbial growth, but also other
inorganic nutrients and cofactors. For the build-up of proteins and nucleic
acids, phosphorus is available in minerals like apatite, while nitrogen predominates most ground waters and is bioavailable via nitrogen fixations34.
18
1.3.4 The start- and end products of the carbon cycle: Methane
and Carbon Dioxide
The greenhouse gases methane and carbon dioxide are of global importance,
as they represent the end products of organic matter decay and are of significance in global warming36. In the scope of the here presented work, they are
especially relevant as these are the central components for chemolithoautotrophic primary production. Methanogenic archaea can synthesize biomass
by using CO2 and hydrogen as carbon and energy sources. In anaerobic hydrocarbon degradation networks, methanogens are obligate partners for the
syntrophic primary degraders, by removing hydrogen that would otherwise
hamper this fermentative process. Biotically, methane is produced enzymatically by methanogenic archaea from a variety of substrates. These are the
three pathways simplified to the chemical reaction37–39:
(i) Hydrogenotrophic methanogenesis
CO2 + 4 H2 à CH4 + 2 H2O
Carbon dioxide is being reduced to methane using hydrogen as the electron
donor.
4 HCOOH à 3 CO2 + CH4 + 2 H2O
Hydrogenotrophs can also utilize formate or secondary alcohols, which substitute for hydrogen.
(ii) Acetoclastic methanogenesis from acetate
CH3COO- + H+ à CO2 + CH4
This reaction occurs in combination with acetate fermentation, in which the
acetate is simultaneously oxidized and reduced (disproportionation reaction)
in order to form carbon dioxide and methane.
(iii) Methylotrophic methanogenesis
4 CH3OH à 3 CH4 + 1 CO2 + 2 H2O
Similarly to the acetoclastic pathway, methylated compounds such as methanol, methylamines and methylsulfides undergo a disproportionation reaction in order to produce methane and carbon dioxide.
Methane oxidation is the reverse reaction in which methane is anaerobically oxidized to carbon dioxide. This still rather enigmatic process has predominantly been previously described for anoxic marine and freshwater
systems. Analogous to methanogenesis, this involves a syntrophic partnership40 using the terminal electron acceptors sulfate, nitrite or metal oxides
such as manganese or iron41.
19
1.4 Metagenomics and the Uncultured World
Microbes organized in communities are strongly interconnected to the surrounding environment and other organisms. In order to understand this complex interplay, cultivation based techniques are not sufficient, as it is well
established that laboratory settings fail miserably in reproducing the myriads
of factors that microorganisms are subjected to42. A consequence of this is
that only a fraction of the microbes that are found in the natural sample can
be cultured, a phenomenon termed “the great plate count anomaly”43.
In the mid 1980s microbiologists came to the realization that they had to
drop the concept that only culturable microorganisms exist. At this point it
was realized that a more systemic approach was needed to understand the
microbial world42. In 1985, Norman Pace was the first to announce the idea
of cloning marker genes directly from the environment44, but it was not until
the 1990s that the first study on cloning the universal taxonomical marker
gene of the 16S rRNA derived directly from marine samples was published45. DNA taken directly from the environment without prior cultivation
of the organisms has been termed the “metagenome” and the study thereof
“metagenomics”46.
The development of “next generation sequencing” (NGS) and the drastic
reduction of sequencing costs, enabled scientists to go beyond the question
of “who is there” when sequencing phylogenetic marker genes, to “what can
they do” by collecting the total genomic information from a given environment. By applying large scale sequencing, it became possible to correlate
genomic content with environmental conditions, to disentangle the relationship between symbiosis partners and to dig deeper into the diversity of a
gene families47, just to name a few possibilities.
By sequencing single cells or pure cultures, the extracted DNA will represent the genomic content of a clonal population of cells (providing there is
no contamination). This will not be the case with a metagenome, where cells
with distanced evolutionary history will be sequenced simultaneously. However, gaining complete coverage of all organisms within a complex sample is
in most cases unattainable, and this derives both from incomplete DNA recovery as well as the difficulty in assembling reads from different but closely related organisms which create interspecies chimeras47.
The identification of genes is mostly based on annotations to known
homologs in databases. This approach is more reliable if a closely related
organism has been sequenced and the function of the encoding genes has
been experimentally verified. BLAST (Basic Local Alignment Search
Tool)48,49 is commonly used to identify gene family members within a metagenome. This approach can be combined with ab initio gene prediction tools
which are mostly based on supervised learning and statistical pattern recognition methods, using Markov or hidden Markov models47,50.
20
In order to assign functions found in the metagenome to the respective
organism, short sequence reads are first assembled to longer fragments (contigs) and subsequently “binned”. One strategy to accomplish this is either by
grouping contigs according to their coverage, GC content, called compositional binning, or phylogenetic binning based on homology searches to reference sequences47.
21
2 The Present Thesis
2.1 Aims
The different chapters in this thesis compare and contrast the effects of distinct environmental gradients (level of connectivity to the surface; hydrocarbon contamination gradient) on microbial communities living in energylimited systems. Using metagenomic approaches, patterns of community
structure, linkages and dependencies could be revealed.
The following main objectives were addressed:
• Chapter I: Reconstructing microbial food webs on the metagenome level:
identifying co-dependencies and partnerships at different degrees of isolation
from the surface. What replaces the sun as a main energy source in deep
aquifers?
• Chapter II: The effects of varying connectivity levels to the sunlit surface
on microbial diversity and community structure. What happens to a microbial community in different water masses over six years?
• Chapter III: Assigning identity to specific metabolic roles of primary alkane degraders and their syntrophic partners beyond transfer of electrons and
intermediate fermentation products. What interactions can we predict among
community members in hydrocarbon degradation under methanogenic conditions?
• Chapter IV: Is there a higher remediation capacity in heavily contaminated
sites when compared to more pristine locations? Why do aromatic pollutants
resist degradation on biologically active sediments?
22
2.2 Methods
2.2.1 Study Sites
Äspö
The Äspö Hard Rock Laboratory (Äspö HRL), operated by the Swedish
Nuclear Fuel and Waste Management Co. (SKB), is an underground research facility located in the southeast of Sweden (Fig. 4). It consists of a
3,600 meter long tunnel going down to a depth of 450 m under the Äspö
island.
The geological formation of this area is referred to as the Fennoscandian
Shield, which is dated to be between 1.6 – 3.1 billion years old. The bedrock
mainly consists of granite and quartz-monzodiorite, which has been fractured by glaciations leading to the formation of various aquifers, which have
been studied extensively in terms of geology, hydrology, and chemistry.
Altogether six water bodies from depths between 170 to 450 m, differing in
water type and age were sampled for the different projects of this thesis.
Baltic Sea
Äspö
HA1327B
KA2162B:1
KF0069A01
Figure 4. The Äspö Hard Rock Laboratory (HRL) sampling site and the three boreholes analyzed in paper II.
Alberta Oil Sand Tailings
The Athabasca oil sands (also called the Athabasca tar sands or Alberta tar
sands) are large deposits of bitumen or extremely heavy crude oil, located in
northeastern Alberta, Canada.
23
Oil sands tailings are byproducts of bitumen extraction from surface mining
and processing of oil sands ores, consisting of a mixture of slightly alkaline
water, sand, clay and residual hydrocarbons. Oil sands operations in northern
Alberta, Canada produce ∼1.3 million barrels of bitumen and ∼262.000 m3
of tailings per day that are deposited into settling basins (tailings ponds).
Tailings accumulate since the producing companies operate under a zero
discharge policy.
Figure 5. The Syncrude Mildred Lake plant.
https://commons.wikimedia.org/wiki/File:Syncrude_mildred_lake_plant.jpg#/media/File:Syncrude_mildr
ed_lake_plant.jpg (Licensed under Public Domain via Commons)
Challenges associated with tailings ponds include the presence of inorganic
and organic contaminants such as metals, salts, petroleum hydrocarbons and
the emission of the greenhouse gases CH4 and CO2. The fine tailings settle
by gravity most rapidly during the first 3-4 years after deposition, forming
mature fine tailings (MFT) which then require decades for significant incremental settling. The samples for this study were taken from the Mildred
Lake Settlings Basin (Fig.5).
Lake Grötingen, Jämtland
Lake Grötingen (62°51’ N, 15°29’ E) is located in Grötingen, along the river
Gimån in central Sweden, and part of the Natura 2000 area of Gimån.
During industrial coal and tar production between 1890-1930, the sewage
from industrial processes was directly discharged into the lake. Consequently, a large area of the lake bottom close to the former factory is still covered
by high amounts of tar residue. Three sampling sites were chosen for this
24
study. The first located at the discharge of the factory, the second 2.5 km
upstream and the third 860 m downstream.
2.2.2 Analytical Methods
Sampling
For chapter I and II, ground waters from six separate boreholes differing in
water age and depth (171- 455 m) were sampled from the Äspö Hard Rock
Laboratory, in the South of Sweden.
The study in chapter III was on short-chain alkane degrading cultures
(SCADC) isolated from mature fine tailings from the Mildred Lake Settlings
Basin in Alberta, Canada.
For chapter IV sediment cores were taken from the tar-contaminated lake
Grötingen located in Jämtland, Central Sweden.
Molecular analysis
The foundation for all studies in this thesis is in depth analysis of recovered
environmental DNA sequences that were interpreted within the respective
environmental context. Each dataset included chemical analyses of the corresponding habitat to examine genetic adaptation to the prevailing conditions.
Single cell picking & whole genome amplification.
To study populations of filamentous Archaea from an alkane degrading consortium (chapter III), single cells were picked under the microscope using a
micromanipulator (Eppendorf TransferMan® NK 2). This mechanical device
enables the uptake and transfer of single cells with a microcapillary and
hence enable recovery of a simplified sub-community from initially complex
samples.
In order to attain a sufficient amount for DNA sequencing, the filaments
were subjected to genome amplification, termed multiple displacement amplification (MDA)51. Random primers bind to denatured DNA followed by
strand displacement synthesis of the Phi 29 polymerase. Each displaced
strand serves additionally as a template, generating high yields of amplified
DNA. The phage derived Phi 29 polymerase, furthermore exhibits proofreading activity that delivers up to 1000-fold higher fidelity compared to the
Taq DNA polymerase52.
Barcoded amplicon sequencing.
Partial 16S rRNA gene sequences were obtained for the studies in chapter II
and III, by using a two-step PCR procedure with primers targeting the V3
and V4 regions of the 16S rRNA gene. The first amplification was to acquire
more material while minimizing amplification bias and amplicons from this
step were subsequently used as template for a second PCR with identical
25
primers, except that both forward and reverse primers included 7 bp DNA
barcodes that were unique for each amplified sample. The obtained
16SrRNA amplicons were submitted to the SciLifeLab SNP/SEQ sequencing facility (Uppsala University) where Illumina MiSeq technology was
used. An in-house sequence analysis and annotation pipeline53 was then applied on the obtained 2 x 300 bp reads, containing following steps: barcode
demultiplexing, read pairing, quality control, clustering and taxonomic assignment against the SilvaMod database.
Whole genome shotgun sequencing.
DNA from the studies conducted in chapters I- IV were submitted to the
SciLifeLab SNP/SEQ sequencing facility (Uppsala University) where metagenome library construction54,55 and Illumina HiSeq sequencing was performed. In short, genomic DNA was sheared using a focused-ultrasonicator,
followed by preparation of sequencing libraries generated for clustered
flowcell sequencing, which was performed on the Illumina HiSeq sequencer
operated in 2 x 100 cycle mode.
Sequence analysis involved trimming, assembly, binning (only for chapters I & III) and gene prediction/annotation. For assembly, MEGAHIT56 (in
chapter II and III) and RAY57 (chapter I) were used. Contigs were scaffolded
with NEWBLER (chapter I) or the BESST software58 (chapter III). Coverage
was computed by mapping back the reads to the scaffolds using bowtie259,
and for computing coverage, bedtools60 was used. After filtering out short
contigs and scaffolds, Concoct61 was used for coverage and nucleotide compositionbased binning. The criteria for high quality bins, so called metagenome-assembled genomes (MAGs), are based on completeness and contamination. An additional criteria used in chapter III was an overlap by more
than 95% in contig content as defined by concoct and taxonomic binning.
We used PHYLOPYTHIA+S62, which uses a marker gene approach to distribute conitigs and scaffolds into taxonomical clades.
2.3 Results
2.3.1 Exploring the Deep Terrestrial Biosphere
The finding that life is possible independent of light energy, resulted in a
paradigm shift in the 1980s63. In fact, it has been estimated that the terrestrial
deep biosphere hosts up to 20% of Earth’s biomass64. Difficulties in accessing these deep terrestrial systems still make them one of the least understood
ecosystems on earth.
26
It has been hypothesized that these ecosystems are sustained by more or
less omnipresent H2 and CO2 and to some extent CH4. All three are viewed
as geo-gases and can be derived from their geological surroundings.
Deprived of high yielding energy sources, we expect to find microbial populations with extremely long generation times stretching from hundred to
thousands of years.
Paper I. The microbial interactome of the terrestrial deep biosphere
reveals metabolic partitioning among populations.
Aim: To infer the metabolic potential and traits of dominant community
members in three deep aquifers with contrasting connectivity to the surface.
Here, we hypothesize that with decreasing connectivity and according energy limitation, cells will be selected to become streamlined.
Figure 6. Model of potential metabolic pathways at depths of -171 m (red), -415 m
(green), and -448 m (blue).
Study: Three aquifers located at depths between 171- 450 m were sampled
for two different cell fractions, one for cells bigger than 0.22 !m and one for
cells and viruses smaller than 0.22 !m. Apart from viruses, the smaller size
27
fraction also contained an abundance of reads assigned to bacteria, therefore
it was included in the study.
The recovered Illumina HiSeq sequences allowed the metabolic reconstruction of 69 distinct microbial genomes, each with > 86% coverage.
Based on metabolic key characteristics, the most abundant populations
from each water mass were classified into broad functional groups (I-VI).
Traits considered were e.g. fermentation pathways, respiration pathways, C
and N fixation. Significant metabolic pathways represented by these functional groups were mapped for each aquifer and a conceptual model of putative metabolic interactions and dependencies were combined into an interactome (Fig. 6).
In the small cell fraction, a phylogenetically distinct subset with reduced
estimated genome size was observed in all three aquifers. This is analogous
to other studies carried out in oligotrophic environments65, showing that in
order to save energy for replicating unnecessary genes, they are lost over
time. A smaller cell size is another adaptation to higher efficiency, by minimizing nutrient needs and increasing the volume to surface ratio and hence
also the capacity to transport solutes65.
With longer isolation from the surface, the communities are depleted in
organic matter that otherwise was washed in by recharge water. This depletion lead to a tendency to depend more on chemosynthesis, which was reflected by the observation of a more heterotrophic or mixotrophic lifestyle
for the upper aquifers, while the deepest and most disconnected site was
dominated by chemolithoautotrophic processes, especially sulfate reduction.
In general, this lifestyle generated by isolation from the surface, created
tight interdependencies between the community members.
Paper II. Connectivity driven bacterial diversity patterns and functional
potential in three deep aquifers of the Fennoscandian shield.
Aim: To characterize the general effects of isolation on the community dynamics and diversity patterns of three deep aquifers over a time period of six
years. Here, we speculate that with increased isolation (disconnection to the
sunlit surface) chemolithautotrophic processes will become more important
in the community and resulting slow growth in combination with long (or
strong) selection has lead to species extinction. This species extinction driven by dispersal and energy limitation should be reflected by a decrease in
overall richness with depth.
Study: In this study, microbial communities of three deep terrestrial subsurface aquifers of depths between 180- 455 m were investigated by sequencing
of 16SrRNA amplicons and shotgun metagenomes. The amplicon dataset
revealed the phylogenetic structure, diversity and community dynamics over
28
six years. Two samples of the deeper aquifers resulted in the successful recovery of whole shotgun metagenomes that were used to detect homologs to
key genes involved in nitrogen and carbon fixation, sulfate reduction, sulfide
oxidation and fermentation.
Figure 7. Bacterial community alpha diversity measures of the three aquifers, showing A) permuted beta-dispersion to test on homogeneity of the samples within the
water masses; B) chao1 alpha diversity estimation; C) Simpson diversity index
(1/D); and D) Simpson evenness (E1/D).
The community at the shallow site, that has in-flow of Baltic Sea water every couple of weeks, showed a highly dynamic population, dominated by
putatively sulfur oxidizing Sulfurovum and Sulfurimonas genera at all time
points. The intermediate aquifer (with a water turnover of approximately five
years) had a less variable and dynamic microbial community and was strongly dominated by candidate phylum OD1. The deepest water mass (5000
years old saline glacial melt and meteoric waters) had the lowest taxon-
29
richness and surprisingly contained Cyanobacteria, that likely were trapped
since the last water in-flow thousands of years ago.
In this study, the degree of connectivity to the surface seems to be the
main factor in shaping community dynamics and microbial diversity.
In accordance to paper I, this study showed that the occurrence of key
genes for nitrogen and carbon fixation, sulfate reduction, sulfide oxidation
and fermentation, point towards a gradient in the ratio between chemoorganohetero- and chemolithautotrophic processes in deep aquifers that is related
to connectivity to the sunlit surface.
We furthermore observed that species richness (Fig. 7 B) declines with
isolation of the environment, which is due to a decreasing number of ecological niches from depleting resources and gradients. Another reason is that
immigration events that would bring new organisms, are more rare.
In contrast, the evenness (Fig. 7 D) inreases with depth. We speculate
that, as species invasions become extremely rare, interactions between remaining community members become tighter and more indispensable under
such a condition. Moreover, preceding succession events could have already
taken place through competition and/or predation leading to an overall lower
richness and to higher evenness, by eliminating dominant species.
2.3.2 Studies on Petroleum impacted environments
Oil derived hydrocarbons are some of the world’s most widespread contaminans and therefore prioritized in environmental management. Environmental-friendly decontamination by microorganisms has in many cases been
demonstrated to be the last resort for removal of these toxic compounds.
Especially biodegradation under anaerobic conditions is a most challenging
task and studying microbial biodegradation of pollutants is important for
several reasons:
1. Finding ways to catalyze and improve bioremediation.
2. The adverse or beneficial effects of microbial activity on hydrocarbon
profiles for the oil industry.
3. The production of the greenhouse gases methane and carbon dioxide at the
end of hydrocarbon degradation. This has an implication from an environmental point of view, but also for methane as a commercial energy resource.
Paper III. Metabolic partitioning in an alkane degrading bioreactor
operating under methanogenic condition.
Aim: To reconstruct the interdependencies within syntrophic microbial assemblages beyond the first level of electron and intermediate metabolite
transfer. Here, we hypothesize that enhanced cross-feeding is needed for
metabolic processes to be optimized for reduced individual metabolic burden
when living in conditions of shortage of bioavailable energy.
30
Study: In order to study the underlining processes of alkane degradation, a
microbial community indigenous to oil sands tailing ponds in Alberta, Canada, was isolated. These cultures were grown under strictly methanogenic
conditions on short-chain alkanes in bioreactors.
The dataset included the full sample of this mixed culture, as well as Archaeal-Bacterial consortia that were isolated using micromanipulation.
Combined with metagenomic approaches, we obtained sequence data of
these synergistic consortia.
CnH2n+2
Secondary degrader
glycolysis
re
c
fu ycli
m ng
ar
a t to
e
b
TCA
CH4
IFM
Acetate
CO2
Vitamins/AminoAcids
oxida
eta
tion
Methanogen
CO2 + H2
Fermentation or
Alkane activation
Primary degrader
produced
& consumed by all
H2
Acetogenic bacteria
Acetate
CO2 + CH4
Methanogen
Figure 8. A proposed model for the syntrophic network and associated secondary
partners.
Three types of syntrophic consortia were identified, each dominated by different types of methanogens, represented by the genera Methanoregula,
Methanosaeta and Methanolinea. Metabolic reconstruction revealed that the
bacteria specifically associated with these methanogens, perform fermentation, as well as a syntrophy and acetogenesis facilitated by energy conservation revolving around H2 metabolism. Other taxa were shown to scavenge on
anabolic products (proteins and lipids) derived from detritus.
The genomic reconstruction of this study points towards more complex
forms of metabolic dependencies, beyond the classic H2-producing,
syntrophic alkane degrader and the methanogenic partner merely scavenging
H2. We rather detected a complex network of strict labor division (Fig. 8).
31
Essential, but metabolically costly biosynthesis functions were lost in various taxonomic groups, indicating that energy and nutrient limitations select
for a streamlined genome (link to paper I). Beyond that, sharing the metabolic burden is an optimization strategy for a community living under thermodynamic conditions that are close to hinder cellular metabolism.
Paper IV. Characterization of microbial communities along a tarcontamination gradient in lake Grötingen.
Aim: Assessing the factors causing the slow bioremediation and to determine differences in functional gene pools at varying pollutant concentrations. Here, we expected tar contaminated sites to have a higher genetic potential for hydrocarbon degradation than pristine sites.
Study: Lake Grötingen became heavily polluted when sewage from industrial production of coal and tar was discharged directly into the lake. Hundred years after the dismantling of this factory, a large area of lake sediment
is still covered with high amounts of tar residue in form of polycyclic aromatic hydrocarbons (PAHs), which are the most problematic to degrade.
We studied differences of microbial bioremediation capacity within three
sites of varying distance to the pollution source. Community composition
(Fig. 9) and genetic potential of microbial tar degraders were studied using
whole metagenome shotgun sequencing. Broadly defined metabolic functions were assessed with MEGAN, while a separate blast search was carried
out for specific marker genes involved in hydrocarbon degradation. Combined with chemical analyses on the lake sediment cores, we furthermore
aimed to assess factors causing the long bioremediation time, especially at
the heavy contaminated site.
32
Community Composition
taxon
upstream3
Acidobacteria
Actinobacteria
Alphaproteobacteria
Armatimonadetes
Bacteroidetes
Betaproteobacteria
Chlorobi
Chloroflexi
Cyanobacteria
Deltaproteobacteria
Epsilonproteobacteria
Euryarchaeota
Firmicutes
Gammaproteobacteria
Gemmatimonadetes
Ignavibacteriae
Nitrospirae
other Archaea
other Bacteria
Planctomycetes
Verrucomicrobia
upstream2
upstream1
downstream3
downstream2
downstream1
outlet3
outlet2
outlet1
0
25
50
relative abundance %
75
100
Figure 9. The microbial community composition of sediment cores taken from three
different sampling sites in Lake Grötigen, Jämtland.
The comparison of functional genes connected to hydrocarbon degradation,
although not statistically significant, indeed pointed towards a slight accumulation of these genes at the site closest to the factory. Overall community
and functional composition did not indicate specific differences between the
sites, although the heavily polluted site did show a distinct community pattern compared to the rest.
The main factor for the slow degradation rate was likely low bioavailability of the pollutants. The fact that we did not observe major differences in
the gene pool implies that the potential to degrade a wide variety of compounds is likely a general feature of microorganisms in lake sediment. Thus
the presence of certain compounds may not have an effect on the inherent
potential, but instead on the transcription rate of the genes involved in this
process.
2.4 Conclusions and Outlook
Any system that is devoid of photons and feature few steep chemical gradients can be seen as close to chemical equilibrium. Without redox reactions,
life would seize to exist. In the context of the research presented in my thesis, various barriers (soil, rock, water or the vessel of a bioreactor) limit or
prevent photon and matter entry at least temporarily. This causes these systems to move towards the thermodynamic limits for life.
33
As a rule of thumb, redox reactions will happen in the order from the energetically most favorable to the reaction with the least energy yield (Fig. 1).
For the deep aquifers this implies, that as long as recharge water supplies the
system with reduced organic matter, nitrates and sulfates, sulfate reduction
and denitrification will dominate. In deeper and more isolated environments
where reduced organic matter is likely more depleted and recalcitrant,
chemolithotrophic processes will instead drive the system. If geo-gases such
as hydrogen and carbon dioxide are available, organisms may depend on
chemolithoautotrophy for primary production. In general these strategies are
energetically less favorable and will result in an overall slower growth rates
within these systems. Consequently generation times of hundred to thousands of years have been suggested for the SLiMES (subsurface lithoautotrophic microbial ecosystems).
In general anaerobic environments, with a rather limited amount of available
free energy, force microbes into tight syntrophic interactions. These complex
syntrophic partnerships require a highly adapted genome streamlined to
maximize energy conservation, a trait in agreement with the observed cellsize minimization. We observed this effect especially in the deep biosphere
and the short chain alkane degrading cultures. In addition, we propose that
cells scavenging anabolic products derived from detrital biomass and intermediate fermentation products are also important in these systems.
As a contrast, the tar contaminated Lake Grötingen serves as an example of a
complex system with substrate alternatives that are more favorable than the
organic pollutants under scrutiny. Tar degrading specialists are either being
outcompeted, or their metabolic plasticity allows them to switch to more
favorable carbon sources, which presumably selects for cells with broad
metabolic capabilities in these communities.
Low energy availability furthermore has an effect on biodiversity. We could
show, that species richness decreased with the grade of isolation from energy
and nutrient sources. Another interesting aspect worth mentioning is related
to species migration and the role of stochasticity. While nutrient and energy
availability are usually the strongest environmental filters, microbes in the
deep biosphere also experience extensive filtering through long time isolation, where predation and the geological formation will hinder major species
invasions. It is perhaps a provocative thought to consider that communities
in isolated biomes are not always capable of exploiting the full spectrum of
available energy and nutrient cycling, simply because functionally essential
members never arrived.
We are still at the beginning of understanding the underlying chemical reactions that enable life on Earth. Furthermore it is likely that each of these
34
reactions are found in various alternative pathways that yet need to be uncovered and described. Metagenomic research is strongly limited by the
incomplete knowledge on metabolic pathways and by the amount of trustworthy assigned functions in databases. Thus research on biochemistry followed by gene expression verification will most certainly remain the foundation for future research in the field.
From a microbiologist’s point of view, I believe it is important to study both
the genome complexity at the level of communities and individual cell (or
filament) if we want to disentangle interactions. A simplified setting, such as
an enrichment culture, allows us to study processes under controlled laboratory conditions, in order to assign function to identity and to understand the
underlying process without cross-signals from peripheral reactions. Studying
genomic potential, especially of streamlined genomes, already tells us a lot
about adaptation to a specific lifestyle. Nevertheless we need to add activity
levels and turnover rates to the equation, which includes a need for transcriptomic and proteomic data.
The deep biosphere still leaves a lot of open questions that need to be asked
and answered. The already mentioned activity measurements need to be
done more extensively on a transcriptome level and linked to actual process
rates, in order to understand the genetic underlying mechanisms of life in
“slow-motion”. Furthermore, studying the trophic effects of viral predation
for shaping the host community and their influence on carbon turnover, as
well as their influence on evolution by facilitating gene-transfer needs to be
put in focus.
In a practical sense, the work on alkane degradation could also help finding
strategies for making recovery of oil sands more efficient by adding essential
metabolites in to the system, that are otherwise energetically expensive for
the microbes.
35
3 Summary in Swedish
Den mörka biosfären breder ut sig under våra fötter! Djupa akviferer i berggrunden, sediment under sjöar och hav och en rad andra liknande miljöer nås
aldrig av det direkta solljus som normalt utgör den huvudsakliga direkta eller
indirekta energikällan för det liv som vi möter i vårt dagliga liv på jordytan.
Av förklarliga, främst praktiska, skäl vet vi fortfarande väldigt lite om de
organismer och processer som kännetecknar dessa svårtillgängliga ekosystem, även om de i såväl volym som biomassa överskuggar de organismer
som lever i solljus. Vad vi vet är att pågående metabola processer och långa
uppehållstider leder till gradvis förbrukning och utarmning av de redoxaktiva ämnen och redox-gradienter som behövs för att driva kemotrof tillväxt och ny produktion av biomassa. Man kan därför betrakta många av
dessa system som vore de i termodynamisk jämvikt där kemiskt bunden
energi för att driva metabola processer är en bristvara.
I mitt avhandlingsarbete har jag utnyttjat storskalig sekvenseringsteknik för
att beskriva artsammansättningen i komplexa mikrobiella samhällen samt
deras samlade och indjviduella genetiskt kodade funktionella potential. Genom att jämföra mikrobiella samhällen i olika typer av miljöer och deras
dynamik över tid har det även varit möjligt att använda denna information
för att beskriva och förstå ekologin hos de organismer och samhällen som
återfinns i dessa miljöer. Det har också varit möjligt att beskriva strategier
och anpassningar för att överleva och tillväxa i miljöer där metabola energiflöden är mycket låga.
Mina huvudsakliga modellsystem har för detta ändamål varit akviferer i den
djupa terrestra biosfären; vattenmassor som rör sig i berggrunden hundratals
meter under våra fötter, och mer ytliga sediment och slaggmassor som förorenats med aromatiska föreningar på grund av mänsklig påverkan. Mina
analyser visar att dessa miljöer gynnar bakterier och arkéer som har reducerade genom och även är små till storleken, vilket skulle kunna vara anpassningar för att spara energi. Som konsekvens saknar dessa organismer ofta
vissa betydelsefulla eller rent av livsnödvändiga gener och funktioner. Att
sakna livsnödvändiga funktioner kan tyckas svårförståeligt, men en uppenbar
fördel är att kostsamma metabola processer kan delas av ett flertal populationer inom ett samhälle. Detta skapar auxotrofier och starka beroendeförhållanden inom samhället. Mina resultat visar också att ytterligare beroendeförhållanden inom den mörka biosfärens mikrobiella samhällen skapas genom
36
att många populationer inte främst utnyttjar primära substrat såsom geo-gas
eller aromatiska kolväten som huvudsaklig energikälla, utan istället lever på
att bryta ner och tillskansa sig sekundära metaboliter eller proteiner, kolhydrater, lipider och andra biomolekyler som producerats vid primärproducenternas biomassatillväxt. Detta pekar på att mikrobiella samhällen som växer
fram i de energibegränsade miljöer som jag studerat är mycket mer komplext
sammansatta än vad som är den gängse uppfattningen och att interaktionerna
går bortom traditionellt definierade anaeroba näringskedjor där gradvis mer
oxiderade metaboliter överförs mellan olika mikrobiella populationer.
I framförallt den djupa berggrundens mikrobiom genomfördes mer detaljerade studier över hur mikrobiella funktioner och populationer varierade mellan vattenmassor och över tid. Även om den huvudsakliga uppbyggnaden av
dessa samhällen med avseende på funktioner var liknande så fanns det också
anmärkningsvärda skillnader. Vattenmassor som under tusentals år varit
isolerade från de solbelysta biosfären uppvisade en mer tydlig hydrogenotrof
och autotrof metabolism, medan mikroorganismer i yngre vattenmassor som
är i mer direkt kontakt med ytvatten är övervägande heterotrofa och sannolikt förlitar sig åtminstonde delvis på tillförsel av substrat därifrån. Det är
kanske inte heller förvånande att dessa mer ytliga bakteriesamhällen uppvisar en betydligt större dynamisk variation över tid, medan bakteriesamhällen
som har en omsättningstid på tusentals år uppvisar en betydligt högre tidsmässig stabilitet och mindre artikedom. Detta kopplar naturligt till förväntad
förbrukning och utarmning av såväl energirika kolföreningar som termodynamiskt fördelaktiga elektronacceptorer över tid, och den sänkta tillväxthastighet som detta i teorin bör leda till.
Situationen i ytliga och höggradigt förorenade sediment ser annorlunda ut.
Här råder ingen direkt brist på näringsämnen eller energirika kolföreningar,
men att bryta ner de stabila aromatiska föroreningar som förekommer i hög
halt är en långsam process som kräver samverkan mellan ett flertal olika
organismer. Här kan tillgången på andra, mer lättillgängliga kolföreningar
vara ett hinder för naturlig nedbrytning av föroreningarna då nedbrytare lätt
konkurreras ut eller styr sin metabolism till att utnyttja mer lättillgängliga
substrat. I de förorenade sedimenten från Grötingen kunde man visserligen
se en avvikande artsammansättning i de mest förorenade sedimenten, men en
vidare analys av de metabola funktioner som uttrycktes, inklusive förekomst
av gener och processer som är direkt involverade i nedbrytning av aromatiska kolväten, visade bara på mindre eller obetydliga skillnader mellan relativt opåverkade och höggradigt förorenade sediment.
Min forskning har således kastat nytt ljus på den hittills dåligt utforskade
mörka biosfären i berggrunden under våra fötter och i sedimenten under våra
ytvatten och förutom att påvisa värdet av funktionell samverkan har faktorer
som styr mikrobiella samhällens sammansättning och funktion sats under
37
lupp. Resultaten kan också användas för att hitta nya vägar för att, genom
exempelvis tillförsel av nyckelmetaboliter eller främjandet av samarbetande
mikrobiella konsortier, styra naturligt förekommande mikroorganismer till
att utföra önskvärda ekosystemtjänster såsom nedbrytning av organiska föroreningar.
38
39
4 Acknowledgements
These past years in Uppsala and Leipzig have been an incredible journey,
not only in a professional way but especially in making me grow personally.
All this would not have been possible without the support of many people
starting with the great triumvirate of brilliant and unconventional minds:
Annelie, Stefan and Alex. Annelie and Stefan, I’m forever grateful for all the
opportunities to work in interesting fields, to cooperate and work internationally and most of all, for your kindness and mentorship over the years.
Alex, thank you so much for guiding and challenging me and at the same
time providing an atmosphere where I could ask stupid questions.
Lars, you are an intimidatingly cool personality and an inspiration. By the
way: sorry for being a nuisance with the study plan every year!
To my cooperation partners Xiaofen & Mark thank you for fruitful discussions and for being a true pleasure to work with!
My past and recent office mates: Birgit, Janne, Tom, Claudia, Hannah &
Jingying. Thank you all for inspiring chats about science and life, especially
Hannah for making me practice my German on a daily basis and helping me
to finish all chocolate within a radius of 30 km J.
Alina (my symbiosis partner in hedonistic-guilty-pleasure-indulgence) and
Frieda (although it took us some time to discover the Amore!)... I have no
words that would give justice to our special friendship! The same goes for
you Inga, you good soul... I will never forget our good times in Germany and
Sweden, and for sure more to come!
Monica, thank you for sharing your home with me for more than three years.
It was a pleasure and thank you for being a good friend and listener.
Moritz and Lucas, you did not only provide crucial groundwork for this thesis, but I especially appreciate you guys for being fun outside of work!
A shout-out to my other fellow PhD’s: Torsten, my favorite rival! Thanks for
the red bull and fixing the graph! Anastasija, the best free-diving and sporadic-climbing partner! Blaize, Jovana, Anne & Roger (ok you guys are a post40
docs by now), Karolina, Lorena, Maria, Yinghua, Martin, Annika, Leyden,
... with all of you I share special memories! Martha and Sari, you incredibly
hardworking and kind people, I can never repay in chocolate or whatever, in
how much you were of help!
Pilar, Omneya, Anna, Andrea, Nuria, Kristin, Dolly, Cristian, Silke, Eva,
Sebastian, Gesa, Peter, (and the ones I forgot to mention)... you all deserve
more than a brief mention; thank you for providing this exceptionally fantastic working environment!
To the Leipzig crew: Karolin, thank you for showing me the beauty of Leipzig and also for your tips and tricks in the lab, which of course goes to Martha and Ute as well!
To those of you who have been with me from the beginning:
Die “Biohendl” Angela, Lisi, Leni, Matsch, Verena, Vero, Mette and
Alex. “Die Gruppe” Lukas, David, Thomas, Andi2, Christoph, Stefan and
Bernadette. Gloria & Hannes, my second home in Berlin. Alice & Martin my
Austrian base in Sweden! I’m incredibly lucky to have you guys as my
friends!
Deep gratitude goes to my family, my foundation. My brother and his family
for their moral support during the past years. To my late father who implanted my interest in biology by getting me the first book I can remember, which
was about dinosaurs (although I eventually ended up studying smaller organisms)! And last but not least I thank my mother, I owe you so much more
than just my existence, but especially for always encouraging me to pursue
my studies.
41
5 References
1. Buchner, P. Endosymbiosis of animals with plant microorganisms. Mycologia
Vol. 60, N, 466–469 (1968).
2. Gilbert, J. A. & Neufeld, J. D. Life in a World without Microbes. PLoS Biol 12,
e1002020 (2014).
3. Edwards, K. J., Becker, K. & Colwell, F. The deep, dark energy biosphere:
intraterrestrial life on earth. Annu. Rev. Earth Planet. Sci. 40, 551–568 (2012).
4. Kieft, T. L., Phelps, T. J. & Fredrickson, J. K. Drilling, coring, and sampling
subsurface environments. ChemInform 39, (2008).
5. Kallmeyer, J. & Dirk, W. Microbial Life of the Deep Biosphere. (De Gruyter,
2014). doi:10.1515/9783110300130
6. Pedersen, K. et al. Evidence of ancient life at 207 m depth in a granitic aquifer.
Geology 25(9):827-, (1997).
7. Stevens, T. O. Subsurface microbiology and the evolution of the biosphere.
Microbiol. Terr. Deep Subsurf. Amy PS, Haldeman DL CRC Press. Boca Rat.
Florida 205–223 (1997).
8. Lovley, D. R. & Chapelle, F. H. Deep subsurface microbial processes. Rev.
Geophys. 33, 365–381 (1995).
9. Kallmeyer, J., Pockalny, R., Adhikari, R. R., Smith, D. C. & D’Hondt, S. Global
distribution of microbial abundance and biomass in subseafloor sediment. Proc.
Natl. Acad. Sci. 109 , 16213–16216 (2012).
10. Sherwood Lollar, B. et al. Hydrogeologic Controls on Episodic H2 Release
from Precambrian Fractured Rocks—Energy for Deep Subsurface Life on Earth
and Mars. Astrobiology 7, 971–986 (2007).
11. Sherwood Lollar, B. et al. Abiogenic methanogenesis in crystalline rocks.
Geochim. Cosmochim. Acta 57, 5087–5097 (1993).
12. Sherwood Lollar, B. et al. Evidence for bacterially generated hydrocarbon gas
in Canadian shield and fennoscandian shield rocks. Geochim. Cosmochim. Acta
57, 5073–5085 (1993).
13. Stevens, T. O. & McKinley, J. P. Lithoautotrophic Microbial Ecosystems in
Deep Basalt Aquifers. Sci. 270 , 450–455 (1995).
14. Kyle, J. E., Eydal, H. S. C., Ferris, F. G. & Pedersen, K. Viruses in granitic
groundwater from 69 to 450 m depth of the Aspö hard rock laboratory, Sweden.
ISME J. 2, 571–4 (2008).
15. Berner, R. A. The long-term carbon cycle, fossil fuels and atmospheric
composition. Nature 426, 323–326 (2003).
16. Kvenvolden, K. A. & Cooper, C. K. Natural seepage of crude oil into the marine
environment. Geo-Marine Lett. 23, 140–146 (2003).
17. Colwell, R. R., Walker, J. D. & Cooney, J. J. Ecological aspects of microbial
degradation of petroleum in the marine environment. CRC Crit. Rev. Microbiol.
5, 423–445 (1977).
18. Head, I. M., Jones, D. M. & Roling, W. F. M. Marine microorganisms make a
meal of oil. Nat Rev Micro 4, 173–182 (2006).
42
19. Coates, J. D., Chakraborty, R. & McInerney, M. J. Anaerobic benzene
biodegradation—a new era. Res. Microbiol. 153, 621–628 (2002).
20. Baek, S. O. et al. A review of atmospheric polycyclic aromatic hydrocarbons:
Sources, fate and behavior. Water. Air. Soil Pollut. 60, 279–300 (1991).
21. Perry, J. J. & Gibson, D. T. Microbial metabolism of cyclic hydrocarbons and
related compounds. CRC Crit. Rev. Microbiol. 5, 387–412 (1977).
22. DeLong, E. F. Microbial Life Breathes Deep. Sci. 306 , 2198–2200 (2004).
23. Morris, B. E. L., Henneberger, R., Huber, H. & Moissl-Eichinger, C. Microbial
syntrophy: interaction for the common good. FEMS Microbiol. Rev. 37, 384–
406 (2013).
24. Schink, B. Synergistic interactions in the microbial world. Antonie Van
Leeuwenhoek 81, 257–261 (2002).
25. Narihiro, T., Nobu, M. K., Kim, N.-K., Kamagata, Y. & Liu, W.-T. The nexus
of syntrophy-associated microbiota in anaerobic digestion revealed by long-term
enrichment and community survey. Environ. Microbiol. 17, 1707–1720 (2015).
26. Díaz, E., Jiménez, J. I. & Nogales, J. Aerobic degradation of aromatic
compounds. Curr. Opin. Biotechnol. 24, 431–42 (2013).
27. Heider, J., Spormann, A. M., Beller, H. R. & Widdel, F. Anaerobic bacterial
metabolism of hydrocarbons. FEMS Microbiol. Rev. 22, 459–473 (1998).
28. So, C. M., Phelps, C. D. & Young, L. Y. Anaerobic Transformation of Alkanes
to Fatty Acids by a Sulfate-Reducing Bacterium, Strain Hxd3. Appl. Environ.
Microbiol. 69 , 3892–3900 (2003).
29. Callaghan, A. V. Enzymes involved in the anaerobic oxidation of n-alkanes:
from methane to long-chain paraffins. Front. Microbiol. 4, (2013).
30. Foght, J. Anaerobic biodegradation of aromatic hydrocarbons: pathways and
prospects. J. Mol. Microbiol. Biotechnol. 15, 93–120 (2008).
31. Callaghan, A. V, Wawrik, B., Ní Chadhain, S. M., Young, L. Y. & Zylstra, G. J.
Anaerobic alkane-degrading strain AK-01 contains two alkylsuccinate synthase
genes. Biochem. Biophys. Res. Commun. 366, 142–8 (2008).
32. Biegert, T., Fuchs, G. & Heider, J. Evidence That Anaerobic Oxidation of
Toluene in the Denitrifying Bacterium Thauera aromatica is Initiated by
Formation of Benzylsuccinate from Toluene and Fumarate. Eur. J. Biochem.
238, 661–668 (1996).
33. Grundmann, O. et al. Genes encoding the candidate enzyme for anaerobic
activation of n-alkanes in the denitrifying bacterium, strain HxN1. Environ.
Microbiol. 10, 376–385 (2008).
34. Pedersen, K. Microbial life in deep granitic rock. FEMS Microbiol. Rev. 20,
399–414 (1997).
35. Falkowski, P. G., Fenchel, T. & Delong, E. F. The Microbial Engines That
Drive Earth’s Biogeochemical Cycles. Sci. 320 , 1034–1039 (2008).
36. Change, C. The Physical Science Basis, Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change,
2007, S. Solomon, D. Qin, M. (2007).
37. Thauer, R. K. Biochemistry of methanogenesis: a tribute to Marjory
Stephenson. 1998 Marjory Stephenson Prize Lecture. Microbiology 144 ( Pt 9,
2377–406 (1998).
38. Ferry, J. G. Methane from acetate. J. Bacteriol. 174, 5489–5495 (1992).
39. Zehnder, A. J. B. Biology of anaerobic microorganisms. (John Wiley and Sons
Inc., 1988).
40. Hinrichs, K.-U., Hayes, J. M., Sylva, S. P., Brewer, P. G. & DeLong, E. F.
Methane-consuming archaebacteria in marine sediments. Nature 398, 802–805
(1999).
43
41. Joye, S. B. Microbiology: A piece of the methane puzzle. Nature 491, 538–539
(2012).
42. Handelsman, J. Metagenomics: Application of Genomics to Uncultured
Microorganisms. Microbiol. Mol. Biol. Rev. 68 , 669–685 (2004).
43. Staley, J. T. & Konopka, A. Measurement of in Situ Activities of
Nonphotosynthetic Microorganisms in Aquatic and Terrestrial Habitats. Annu.
Rev. Microbiol. 39, 321–346 (1985).
44. Pace, N., Stahl, D., Lane, D. & Olsen, G. The analysis of natural microbial
populations by ribosomal RNA sequences. Adv. Microb. Ecol. SE - 1 9, 1–55
(1986).
45. Schmidt, T. M., DeLong, E. F. & Pace, N. R. Analysis of a marine picoplankton
community by 16S rRNA gene cloning and sequencing. J. Bacteriol. 173, 4371–
4378 (1991).
46. Handelsman, J., Rondon, M. R., Brady, S. F., Clardy, J. & Goodman, R. M.
Molecular biological access to the chemistry of unknown soil microbes: a new
frontier for natural products. Chem. Biol. 5, R245–R249 (1998).
47. Wooley, J. C., Godzik, A. & Friedberg, I. A primer on metagenomics. PLoS
Comput. Biol. 6, e1000667 (2010).
48. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local
alignment search tool. J. Mol. Biol. 215, 403–10 (1990).
49. Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25, 3389–402 (1997).
50. Kunin, V., Copeland, A., Lapidus, A., Mavromatis, K. & Hugenholtz, P. A
Bioinformatician’s Guide to Metagenomics. Microbiol. Mol. Biol. Rev. 72 ,
557–578 (2008).
51. Lasken, R. S. Single-cell genomic sequencing using Multiple Displacement
Amplification. Curr. Opin. Microbiol. 10, 510–6 (2007).
52. Dean, F. B., Nelson, J. R., Giesler, T. L. & Lasken, R. S. Rapid amplification of
plasmid and phage DNA using Phi 29 DNA polymerase and multiply-primed
rolling circle amplification. Genome Res. 11, 1095–9 (2001).
53. Sinclair, L., Osman, O. A., Bertilsson, S. & Eiler, A. Microbial community
composition and diversity via 16S rRNA gene amplicons : evaluating the
Illumina
platform.
PLoS
ONE
10(2)
e0116955.
doi
10.1371/journal.pone.0116955 (2015).
54. Lundin, S., Stranneheim, H., Pettersson, E., Klevebring, D. & Lundeberg, J.
Increased throughput by parallelization of library preparation for massive
sequencing. PLoS One 5, e10029 (2010).
55. Borgström, E., Lundin, S. & Lundeberg, J. Large scale library generation for
high throughput sequencing. PLoS One 6, e19119 (2011).
56. Li, D., Liu, C.-M., Luo, R., Sadakane, K. & Lam, T.-W. MEGAHIT: An ultrafast single-node solution for large and complex metagenomics assembly via
succinct
de
Bruijn
graph.
Bioinforma.
(2015).
doi:10.1093/bioinformatics/btv033
57. Boisvert, S., Raymond, F., Godzaridis, E., Laviolette, F. & Corbeil, J. Ray
Meta: scalable de novo metagenome assembly and profiling. Genome Biol. 13,
R122 (2012).
58. Sahlin, K., Vezzi, F., Nystedt, B., Lundeberg, J. & Arvestad, L. BESST-efficient scaffolding of large fragmented assemblies. BMC Bioinformatics 15,
281 (2014).
59. Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat.
Methods 9, 357–359 (2012).
44
60. Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for
comparing genomic features. Bioinformatics 26, 841–842 (2010).
61. Alneberg, J. et al. Binning metagenomic contigs by coverage and composition.
Nat Meth 11, 1144–1146 (2014).
62. Gregor, I, Dröge, J. Schirmer , M., Quince, C. & M. A. C. PhyloPythiaS+: A
self-training method for the rapid reconstyruction of low-ranking taxonomic
bins from metagenomes. ArXiv:1406.7123 (2014).
63. Ghiorse, W. C. & Wilson, J. T. Microbial ecology of the terrestrial subsurface.
Adv. Appl. Microbiol. 33, 107 (1988).
64. McMahon, S. & Parnell, J. Weighing the deep continental biosphere. FEMS
Microbiol. Ecol. 87, 113–20 (2014).
65. Giovannoni, S. J. et al. Genome streamlining in a cosmopolitan oceanic
bacterium. Science 309, 1242–5 (2005).
45
Acta Universitatis Upsaliensis
Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology 1301
Editor: The Dean of the Faculty of Science and Technology
A doctoral dissertation from the Faculty of Science and
Technology, Uppsala University, is usually a summary of a
number of papers. A few copies of the complete dissertation
are kept at major Swedish research libraries, while the
summary alone is distributed internationally through
the series Digital Comprehensive Summaries of Uppsala
Dissertations from the Faculty of Science and Technology.
(Prior to January, 2005, the series was published under the
title “Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology”.)
Distribution: publications.uu.se
urn:nbn:se:uu:diva-264115
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2015