Ecology, Microbial☆

Allan Konopka, Purdue University, West Lafayette, IN, United States

r 2019 Elsevier Inc. All rights reserved.

Abbreviation PCR polymerase chain reaction

DOC dissolved organic carbon RTFs resistance transfer factors
HGT horizontal gene transfer SIMS secondary ion mass spectrometry
HPK histidine protein kinase

Glossary Microenvironment The chemical, physical, and biological

Biogeochemistry The study of processes and reactions factors that are located within a few micrometers of the
that govern the composition of the natural environment exterior surface of a microbial cell.
and the cycles of matter and energy that transport the earth’s Oligotrophic An aquatic ecosystem that has low nutrient
chemical components in time and space. resource levels, and as a result, low biomass levels.
Eutrophic Enriched with inorganic nutrient resources, Planktonic Free-floating microbiological organisms
such as N or P, to the level that excessive amounts of dispersed in solution within an aquatic habitat.
phototrophic biomass is produced. The term is primarily Subsurface The zone below the surface of the earth. For
applied to aquatic ecosystems. ocean systems, depths 4200 m may be considered the
Functional redundancy A property of a microbial subsurface. In terrestrial systems, sediment at depths below
community in which different species have the potential to the rooting zone of the plant community may be considered
perform the same functional role in an ecosystem. A the subsurface.
consequence is that reductions in species diversity may not Thermodynamically favorable reaction A reaction in
affect ecosystem functionality. which the Gibbs free energy in the products is less than in
In situ Located in the natural environmental habitat. the reactants. That is, the reaction releases energy. These
Limiting nutrient A nutrient resource whose reactions can occur spontaneously, although the reaction
concentration or flux rate limits the amount of biomass in rate may be greatly enhanced by a catalyst such as an
the habitat, the growth rate of organisms in the habitat, enzyme.
or both.

Defining Statement

This article provides an overview of the field of microbial ecology by delineating its principles, historical development, areas of
study, and experimental approaches.

What is Microbial Ecology?

The term ecology was devised by Ernst Haeckel from the Greek oekologie meaning ‘the study of the household of nature’ in 1866. Hence,
microbial ecology concerns the interactions of microbes with both their abiotic and biotic environments. Microorganisms are ubi-
quitous – they grow not only in soil, freshwater and marine habitats, and on or within plants or animals, but also in
so-called extreme environments with chemical and physical conditions that exclude most if not all other forms of life. Thus, the field is
extraordinarily broad; a number of other articles in this encyclopedia discuss specific topics within this broad field. This article represents
a synthesis of developments that may have arisen from the study of different environments or different organisms, and puts them into a
historical context of how microbial ecology has developed and (more importantly) where it may be heading in the future.
The definition of microbial ecology is very important, because it leads to some important points regarding the dis-
cipline. Microorganisms (by definition) are too small to be seen with the naked eye; bacterial, archaeal fungal, and
protistan cells typically have diameters between 1 and 100 mm, with bacterial sizes heavily skewed to the smaller end of this
distribution and protists at the larger end. Consequently, the ‘environment’ that these organisms are sensing must also be at
a micron scale (the microenvironment). This scale is far too small for humans to sense directly. We can analyze larger,

☆
Change History: June 21, 2018.

Encyclopedia of Microbiology, 4e doi:10.1016/B978-0-12-809633-8.90675-7 1

2 Ecology, Microbial

macroscopic samples and infer what occurs at the microscale, although there is a great risk that the properties we measure
in the large sample merely represent the average of a rich heterogeneity at the microscale. Because of this, technological
innovations are important to microbial ecology when they allow authentic analysis of the in situ environment. The
definition of microbial ecology also indicates that the intellectual interests and approaches of the microbial ecologist are
quite distinct from those of the laboratory-oriented microbiologist who focuses on analysis of pure cultures of microbes.
Microbes in nature interact not only with their abiotic environment but also within biological communities of other
microbes and perhaps plants and animals. These complex systems may exhibit behaviors that are not predictable from
study of the individual organisms in isolation.
That being said, most microbial ecologists have been trained as laboratory-oriented microbiologists rather than as ecologists; as
a result, their understanding of ecological theory is often rudimentary. This can lead to conceptual problems, either because
problems in microbial ecology do not become well integrated into ecological theory or because there is no clear recognition where
the ecology of microbes has elements that are unique from those of plants or animals.

Historical Development of Microbial Ecology

The realization that there was a microbial world depended upon technological innovation – Anton van Leeuwenhoek improved
techniques in lens grinding that led to his visualization of ‘animalacules’ in the late 1600s. However, the discipline of micro-
biology did not really develop until the late nineteenth and early twentieth centuries when Louis Pasteur and Robert Koch took on
practical problems that arose from the interaction of microbes with their environments (such as the germ theory of disease). Their
studies were not ecological, but they did develop technologies for the fundamental manipulation of microbes that are still used
today. Also in the early twentieth century, Martinus Beijerinck and Sergei Windogradsky became interested in the activities of
microbes in natural habitats. They took into consideration the chemical conditions in the natural environment and devised
selective conditions (enrichment cultures) by which they could isolate organisms that carried out specific biogeochemical activ-
ities. However, their focus was not on the habitat per se but on the physiology of the microorganism, and their studies did not
focus on ecological interactions.
Modern microbial ecology had its origins in the 1950s, in the study of the rumen ecosystem by Robert Hungate and his
colleagues. Hungate is well known for his development of techniques for the cultivation of strict anaerobes. In addition, he had an
interest in quantitative analysis of microbial ecosystems. This followed from his studies with C.B. van Niel on cellulose hydrolysis
by gut microbiota first in beetles and termites and subsequently in the bovine rumen. He pioneered studies in which the analysis
did not end at the isolation and characterization of pure cultures, but necessarily included the enumeration of organisms in the
habitat and quantification of their activities. This quantitative analysis distinguished major from minor microbial species and
catabolic pathways in the rumen ecosystem.
As in other areas of science, progress in microbial ecology has been intimately tied to the development of new technologies. The
availability of radioisotopes beginning in the 1950s was a major impetus for the measurement of biogeochemical process rates in nature. A
major problem is that samples must usually be removed from the habitat and incubated under artificial conditions. As incubation time
lengthens, the risk of bottle effects that produce results different from in situ rates increases. Thus, there is a premium on sensitive detection
methods in which incubation time is minimized. The use of 14CO2 allowed analysis of rates of primary production by phototrophs and
chemoautotrophs, and both 14C- and 3H-isotopic forms of a broad variety of organic compounds have been applied to analyze
fundamental processes such as nutrient uptake and assimilation, mineralization to CO2, and microbial community growth rates. 32PO4
and 35S isotopes have been applied to measure biogeochemical process rates in the phosphorus and sulfur cycles. Unfortunately, no
practically useful N radioisotope is available, although advances in the sensitivity of stable isotope analyses may make 15N analyses at short
incubation times feasible. Another significant breakthrough in analysis of microbial process rates came in the 1980s, when microelectrodes
sensitive to certain chemical species were developed. These electrodes have a spatial resolution of 50–100 mm, and have produced great
insights into the spatial and temporal dynamics of microbial processes in habitats such as microbial mats and sediments.
Quantitative analyses of biogeochemical process rates blossomed in the period after radioisotopes became widely
available, but these studies were incomplete because the specific microbial catalysts were not identified. The techniques of
pure culture microbiology were broadly recognized to be ineffective in cultivating numerically abundant organisms from
nature. However, over the past 25 years, a set of cultivation-independent techniques have evolved to determine the relative
abundance of microbes in a habitat. Carl Woese demonstrated that the sequence of the small subunit ribosomal RNA
molecule could be used as a molecular chronometer to analyze the phylogenetic relationships among all forms of life. Norm
Pace took this seminal idea and applied it to the analysis of ‘who’s there’ in natural environments. That is, by isolation of
nucleic acids directly from a natural environment, amplification of small subunit rRNA gene sequences by polymerase chain
reaction (PCR), DNA sequencing of the amplicons, and comparison of the retrieved sequences to the ever-growing database
of sequences from pure cultures and environmental DNA, it was unnecessary to cultivate the organism to determine their
presence. This has become the major activity in the field of microbial ecology over the past 15 years, and has provided
tremendous insights into the phylogenetic diversity present within microbial habitats. However, the determination of what
microbes are present in a habitat does not directly resolve the problem of specifically linking specific microbes to their
quantitative biogeochemical role in habitats. Metagenomics, the sequencing of total DNA recovered from an environment,
now provides insights into biogeochemical potential, but measuring actual expression of that potential remains a difficult
technical problem in the discipline.
 Ecology, Microbial 3

The Scope of Microbial Ecology

Microbes are present in almost all aquatic and terrestrial environments, and associated with all plant and animal species. Although
each of these habitats may have unique aspects that select for particular strategies among microbes, in this general discussion of
microbial ecology, the focus will be on elements that are common to different physical habitats.

Microbes’ impact at different spatial scales

The biosphere can be construed as a series of interlinked systems that operate at different spatial scales. Microbes physically operate at a
micrometer scale, but both impact and are impacted by processes that operate at larger scales. At the largest scales (global and
landscape) are aquatic and terrestrial biomes. Biomes are large geographical areas defined by macroecologists on the basis of their
distinctive flora and fauna, which presumably reflect selection and adaptation to climate and geography. For example, terrestrial biomes
include temperate deciduous forests, grasslands, rainforests, deserts, taiga, and tundra. Marine aquatic biomes would include polar,
temperate, and tropical oceans, and freshwater biomes comprise flowing water (rivers and streams), lakes, and ponds, or wetlands.
Whereas biogeography (the study of the distribution of organisms over space and time) is well established for plants and animals, the
concept has been less intensively studied for microbes until very recently. There are studies in which there is a correlation between
geographic distance and genetic distance for specific bacterial groups. In addition, an increase in the number of taxa detected with
increases in sampling area has been found for microbes, as has been found many times for plants and animals. However, the underlying
mechanisms responsible for the species-area relationship may be quite different in microbes, due to unique mechanisms for phylo-
genetically broad genetic transfer and (in some cases) high rates of recombination.
At the global and landscape scale, biotic and abiotic planetary processes can impact each other across biomes. For example,
oceanic photosynthesis may be limited by the amount of iron; iron is a cofactor for electron transport proteins essential for
conversion of solar energy to a biologically useful form of energy. Arid conditions in deserts produce iron-rich dust particles that
can be transported thousands of kilometers through the atmosphere and deposited in the ocean. As another example, the global
dispersal of animal viruses such as West Nile virus and influenza virus H5N1 can be mediated by migratory birds.
At the ecosystem scale, there are rich, complementary interactions between physical, chemical, and biological components. For
example, photosynthetic sulfur bacteria may occur in freshwater lakes; they require light energy, sulfide as an electron donor that
will be used in large part to reduce CO2 to the oxidation level of cell organic matter, and anaerobic conditions. This combination
of factors only occurs in lakes where morphometry, nutrient inputs, and biology are constrained within certain limits (Fig. 1).
However, to understand the mechanistic basis for interactions of a microbe with its environment, one must consider the conditions
within a few microns of the cell surface, the microenvironment (Fig. 2). The cell’s surface interacts with the external physical environment –
this includes physical factors such as temperature or light but may also involve interactions with the surface of minerals or microbial or
metazoan cells. A broad suite of chemicals are present – some are required nutrient resources, but others may be toxic or represent signals

Fig. 1 System interactions affect the occurrence and intensity of photosynthetic bacteria in lakes. These microbes will only be active at strata
where light, anaerobic conditions, and reduced S cooccur. The diagram depicts important physical, chemical, and biological factors, and whether
an increase in intensity of one factor produces an increase ( þ ) or decrease (  ) in a related factor. For example, increases in inorganic nutrient
concentrations stimulate phytoplankton growth. This has a negative affect on water transparency at depth (necessary for photosynthetic bacteria)
but also results in a larger flux of organic matter to sediments. This stimulates sulfide production, which is also required for photosynthetic
bacterial production. The morphometry of the lake basin and its degree of wind protection drive the intensity of lake mixing, which affects the
depth from the sediment to which anaerobiosis can extend. Reproduced from Parkin, T.B., Brock, T.D., 1980. Photosynthetic bacterial production
in lakes – the effects of light intensity. Limnology and Oceanography 25, 711–718.
4 Ecology, Microbial

Fig. 2 The microenvironment. The volume within a few micrometers from the microbial cell surface has the greatest impact upon it. External
resources are transported across the cell membrane via membrane-bound permeases. Microbial cells contain sensors at their cell surfaces that can
detect chemical signals in the environment. This can lead to changes in gene expression and alterations in microbial cell activity. Reproduced from
Konopka, A., 2006. Microbial ecology – searching for principles. Microbe 1, 175–179.

arising from other cells. The chemicals may arrive at the cell surface by diffusion from the bulk phase liquid, or via intimate contact in
dense biofilms. For nutrient resources, there are membrane-bound permeases that transport molecules into the cell. As single-celled
organisms, microbes have evolved a series of sensors on cell surfaces to detect and respond to chemical cues. One such system results in a
molecular memory of past chemical concentrations and controls the rotation of bacterial flagella in a way that effects chemotaxis. In N2-
fixing root nodule symbioses, chemical cues are passed between Rhizobium species and specific plant legumes; flavonoids excreted from
plant roots are the key signals in initiation of nodule formation in the nitrogen-fixing symbiosis. They bind to NodD proteins in the
rhizobial cytoplasmic membrane and these sensors also activate transcription of nodulation genes critical for the initiation of root nodule
formation. Some of these nod genes specify enzymes that synthesize lipochitooligosaccharides, molecules that are sensed by root cell
sensors and lead to morphological changes in them. Environmental sensors may also be important in animal pathogenesis. The low pH of
the human stomach is a defense mechanism against bacterial infection of the gastrointestinal tract. In the human pathogen Vibrio cholerae,
however, transcription of genes required for colonization and pathogenesis is induced by exposure to acid and higher temperatures as
would occur after a host ingested the organism from the environment.
A phylogenetically widespread regulatory system that can detect environmental signals and produce changes in bacterial gene
expression is two-component regulatory systems, in which one component is a histidine protein kinase (HPK). Approximately
5000 have been identified by bioinformatic analysis of more than 200 bacterial genomes. The proportion of the genome that
encoded HPKs increased as genome size increased, and many of the organisms with very high proportions of HPK exist in
anaerobic sedimentary habitats in which there are steep chemical gradients over short scales (such as δ- and e-proteobacteria) or
have relatively complex life cycles (the cyanobacterium Nostoc and aquatic chemoheterotroph Caulobacter crescentus). On the other
hand, bacteria that are intracellular parasites of eukaryotes might be expected to encounter a more constant external environment
and they were found to have few HPK genes.

Categorizing microbial ecosystems

A focus on the microscopic scale suggests a set of microbial ecosystems (Table 1), based upon the relevant physical and chemical
characteristics at that scale. The quintessential characteristic of aquatic habitats is the planktonic (free-floating) lifestyle. Growth of
planktonic chemoheterotrophic bacteria is generally limited by labile, dissolved organic carbon (DOC) present as sugars, amino acids,
and organic acids that arise from larger organisms via excretion, lysis, or hydrolysis of macromolecules. The total DOC concentration in
ocean waters may approach 1 mg l1, but this represents a pool of several hundred different organic molecules, each present at very low
(nmol l1 to mmol l1) concentrations. The concentrations are low because the bacteria have evolved high-affinity transport systems,
and competitive forces select for microbes that can exploit multiple substrates at low concentrations.
Other microbes must deal with strong concentration gradients of chemical substances, because they are relatively fixed in a solid
matrix (such as an aquatic sediment, subsurface aquifer, microscopic biofilm, or macroscopic microbial mat). Strong spatial gradients
form in these systems because the chemical sources and sinks differ in location. If microbes are sinks or sources for chemical species,
their numbers and activity will affect the steepness of the gradients. These gradient systems may be stable in time or show significant
temporal fluctuations. For example, microbial mats contain photoautotrophic cyanobacteria in the top few millimeters of the mat and a
variety of chemoheterotrophic bacteria are located deeper. Experiments with microelectrodes have shown that there are large and rapid
changes in chemical species such as O2, H2S, and H þ just after the sun rises and after it sets (Fig. 3).
Terrestrial soil habitats that are not saturated with water may be likened to vast deserts in which there is an occasional
convergence of water, nutrients, and microbial cells on a particle that results in a burst of microbial activity. Soils may seem to be
 Ecology, Microbial 5

Table 1 Microbial-scale ecosystems

Ecosystem type Examples Characteristics

Planktonic Open ocean, lakes ‘Oligotroph lifestyle’ High affinity for uptake of multiple
nutrients
Surface-associated Freshwater and ocean sediments, subsurface sediments, ‘Gradient lifestyle’ Hydrodynamic processes and fluid flow
saturated water microbial mats, biofilms determine nutrient fluxes
Biomass density affects gradient steepness
Surface-associated Surface and vadose zone soils Water availability as limiting factor for activity and dispersal
unsaturated water
Patchy nutrient distribution Dormancy
Macroorganism associated Gastrointestinal tract, rhizosphere, epiphytes Coevolution, specific molecular interactions with surface-
associated molecules

Reproduced from Konopka, A., 2006. Microbial ecology – searching for principles. Microbe 1, 175–179.

Fig. 3 Microbial interactions and dynamics in a microbial mat. The flow diagram illustrates major functional groups of microbes and
transformations in the biogeochemical C and S cycles. The depth profiles on the left present idealized gradients of O2 and H2S during the day
(when photosynthetic oxygen evolution occurs) and at night (when respiratory processes lead to anoxic conditions through much of the mat).
Reproduced from DesMarais, D.J., 2003. Biogeochemistry of hypersaline microbial mats illustrates the dynamics of modern microbial ecosystems
and the early evolution of the biosphere. The Biological Bulletin 204, 160–167.

rich with microbes (109 bacterial cells and thousands of different bacterial genomes per gram), but these numbers occupy o0.01%
of the soil volume. The spatial location of microbial activities in soil can be conceptualized as a series of three-dimensional maps
that overlay each other within a matrix of pores. These maps describe the spatial location of (1) different microbes (which have
distinct biochemical functionalities), (2) chemical resources (organic and inorganic), and (3) water and gases. Each is necessary
but none by themselves is sufficient – activity will proceed only where appropriate levels of all necessary factors intersect. The result
is a complex three-dimensional landscape in which the physical structure, determined by not only the mineral matrix but also by
plant roots, microbial cells, and extrapolymeric substances, sets the conditions for the activity of microbes and the fate of organic
and inorganic C, N, and other elements.
In the case of plants or animals as microbial ecosystems, there is an additional dimension to the chemical gradients and
physical structure of the ecosystems described above. That is the capacity for very specific molecular interactions with biomolecules
on cells of the plant or animal, and the capacity for (co)evolution by both the microbe and its ‘environment’.

Thinking like a microbe

Microbial activity at the microscopic system level produces effects at the macroscopic and landscape level. These effects have
relevance for scientific disciplines such as macroecology, biogeochemistry, and geology. However, there are principles unique to
the ecology of microbes, and they only emerge from ‘thinking like a microbe’; that is, by considering interactions with the
environment at the microscopic scale. It is important to keep in mind some characteristics that distinguish microbes from
macroorganisms:
6 Ecology, Microbial

• Small size – Optical amplification (microscopy) is an invaluable tool to assess the distribution of microbes in natural habitats.
• Fast rates of growth and metabolism – Per unit biomass, microbes grow and metabolize much more rapidly than macro-
organisms. Generation times of 1–2 days in natural environments are not unusual.
• Exponential growth of populations – The growth of individual microbial cells is autocatalytic; a cell grows to form two cells,
each of which can catalyze further growth. When coupled with the relatively fast growth rates of microbes, favorable envir-
onmental conditions can produce explosive increases in microbial populations over short period of times.
• Rapid dispersal can occur over long distances – Due to their small size, microbes can be carried for long distances by air or
water turbulence. In addition, animals may transport bacteria via mechanisms such as fecal transmission.
• Genetic flexibility – Although Bacteria, Archaea, and many fungi lack sexual recombination mechanisms, many microbes
possess genetic elements that mediate vectorial gene transfer from donors to recipients. Analysis of microbial genomes shows
that gene transfer is not limited to within a species, but may occur across broad phylogenetic boundaries. In the cases of
resistance to newly introduced antibiotics and metabolism of novel xenobiotic organic compounds, evolution has occurred
over a period of decades rather than millennia.
• Adaptability to environmental extremes – Most microbial species tolerate and grow within what we consider the normal range
of environmental factors such as temperature, pH and chemical concentrations. However, species that are optimally adapted to
these environmental extremes have evolved.

Principles of Microbial Ecology

A statement of general principles is difficult, because the attention of microbial ecologists is diffused over many different habitats.
Thus, there is a tendency to focus on the idiosyncratic rather than the general. In this section, an attempt is made to synthesize
phenomena and search for fundamental principles among the unique features of diverse ecosystems.

Overarching Principles
Principle 1. The primary role of microbes in the biosphere is as catalysts of biogeochemical cycles. That is, they mediate kinetically
inhibited but thermodynamically favorable reactions. Some microbes may also have strong effects on the ecology of specific
micro- or macroorganisms, via their roles as pathogens or symbionts.
Solar energy is the external energy source that drives the biosynthesis of organic matter from inorganic C, N, P, S, and trace
elements by photosynthetic primary producers. These inorganic nutrient sources are present in finite amounts; microbes are
often largely responsible for catabolism of organic compounds by which the mineral forms used by primary producers are
regenerated.
In terrestrial ecosystems, plants are responsible for the great bulk of primary production. Plant organic matter becomes
available for microbial decomposition in forms that vary in molecular structure, susceptibility to degradation, and temporal
availability. These include root exudates, the fine meshwork of root tissue in soil, and the seasonal fall of plant litter onto the soil
surface. Each source may contain complex mixtures of organic substrates that range from low molecular weight sugars, organic
acids, and amino acids to macromolecules. Macromolecule catabolism involves initial hydrolysis to monomers by microbial
extracellular hydrolases and subsequent intracellular metabolism. Not only is organic carbon oxidized back to CO2, but cata-
bolism converts organic N, P, and S to NH4 þ , PO43, and SO42. In this process, energy is released by breaking C–C covalent
bonds; some of the released energy is conserved by microbes and used to drive their growth.
The cycling of organic C in soil intersects with biogeochemical reactions in which inorganic chemical species serve either as
terminal electron acceptors or as electron donors. These reactions are particularly important with respect to N, because these
transformations affect the bioavailability of inorganic N for plants. For example, NH4 þ is a reduced form of N that is used as an
energy source by nitrifying bacteria, and NO3 is the end product of nitrification. However, NO3 is more mobile than NH4 þ in
soil solution, and hence more susceptible to loss. Furthermore, if the soil becomes anoxic, NO3 can substitute for O2 as a
terminal electron acceptor for chemoheterotrophic microbial metabolism. The oxidized end products, N2O and N2, are gaseous
and thus represent a loss of N for plants.
In addition to their biogeochemical function in nature, the specific interaction that microbes have with plants or animals can
have major consequences, not only on the affected organisms but also on the ecosystem. Frank pathogens of plants and animals
can generate selection pressures to evolve resistance determinants upon their hosts. For example, humans with the sickle cell trait
(one allele of an altered form of hemoglobin along with one normal allele) are more likely to survive disease caused by the
malaria protozoan parasite than individuals homozygous for normal hemoglobin. The incidence of sickle cell trait in humans is
highly correlated with areas in which malaria is endemic.
Microbes can also positively affect macroorganisms via mutualisms. A well-studied example relates to another aspect of the
nitrogen biogeochemical cycle. Bacteria such as Rhizobium and Frankia can form specific associations within nodules on plant
roots. These associations make photosynthetically produced organic C available to the microbes. When in association with plants,
these microbes express nitrogenase, an enzyme that fixes N2 into NH4 þ ; microbial metabolism is regulated such that 490% of
fixed N is translocated to the plant.
 Ecology, Microbial 7

Principle 2. Although generally unseen to the naked eye, microbes comprise more than 25% of all biomass on Earth. All
habitats suitable for plants and animals also harbor microbial populations. In addition, some microbes are adapted to grow under
physical and chemical conditions that are too extreme for plant and animal growth.
Determining the amount of microbial biomass on a global scale is difficult to do with precision, but estimates suggest that the
amount of organic carbon in Bacteria and Archaea (ca 200 Pg) is approximately one-third of that found in land plants. Estimates
of eukaryotic microbial biomass (e.g., fungi in terrestrial habitats) are difficult to obtain, but would certainly augment this estimate
significantly. more than half of the biomass on Earth is likely to be microbial. Near-surface terrestrial and aquatic habitats make
larger contributions to overall microbial growth and activity. There are few measurements of growth rate in the deep subsurface,
but existing measures suggest a generation time for microbes measured in tens to thousands of years; this may reflect occasional
periods or locations of microbial growth with most of the microbes surviving in a metabolically quiescent state.
Microbes associated with plants or animals can exert strong ecological effects via parasitism or mutualism. However, in terms of
global biomass, these organisms are minor players; they account for less than 1/100,000th of microbial biomass on Earth. Bacteria
that function in digestion and nutrition of animal herbivores (ruminants and termites) comprise the largest proportion, due to the
large biomass of these animals on Earth.
Humans characterize habitats in which the physical or chemical conditions preclude the existence of plants or animals as
‘extreme’. However, some microbes have evolved adaptations in molecular structure such that they are optimally adapted to
growth in a particular extreme environment such as very acidic or basic pH, very cold or hot temperatures, or high salt con-
centrations. These adaptations usually mean that the organisms cannot grow under the conditions we humans consider normal. In
the case of thermophiles and psychrophiles, evolution in cellular macromolecules has occurred. For example, cellular enzymes in
thermophiles retain structure and activity at higher temperatures; this may be a consequence of more extensive secondary structure
and reduced content of certain labile amino acids. A significant problem at low temperature is fluidity of membranes, and the
lipids in psychrophiles have increased content of fatty acids that will remain fluid at lower temperatures.
Principle 3. The effects of microbial catalysis (the rates of chemical and energy flow through an ecosystem) depend upon the
population size of microbes (the number of catalysts) and their physiological activity (as determined by extant physical and
chemical conditions). In stable ecosystems, small numbers of microbes can still produce significant effects over geological
time spans.
Microbial population levels and cell-specific activities can range over many orders of magnitude in nature. In aquatic eco-
systems such as lakes and coastal marine areas, excessive nutrient enrichment leads to undesirable ecological effects (eutrophi-
cation). The amount of Pi loaded into an aquatic ecosystem will often determine the amount of photoautotrophic biomass (Fig. 4)
in the water column, and systems with higher levels of photoautotrophic biomass exhibit increased levels of primary productivity.
However, physiological studies of phototrophs have shown that the cell-specific rate of photosynthesis will vary significantly as a
function of irradiance, and that even at saturating irradiances the photosynthetic rate may vary 5- to 20-fold, depending upon the
nature and stringency of nutrient limitation.
The effects of temperature upon activity may be even more striking, when one realizes that much of the biosphere (the deeper
waters of oceans and lakes, and many subsurface terrestrial systems) has ambient temperatures of o10 1C. As microbes are
exposed to temperatures near their minimum, not only do the catalytic rate of enzymes decline, but membrane lipids stiffen,
thereby decreasing the effectiveness of membrane-bound transport proteins. As a result, substrate affinity declines and cells are less
effective in concentrating substrates from dilute environments. The physiological consequences are that limiting resources can have
even more acute affects on activity in cold habitats.
Some processes may occur very slowly, yet have significant impacts on biogeochemical cycles. For example, the anaerobic
oxidation of methane in deep-sea sediments is biogeochemically important, given the vast expanse of seabed floor. However, the

Fig. 4 The Vollenweider model of the relationship between the amount of P that enters a lake (P loading in mg P per m3 of lake volume) and the
standing crop of phytoplankton biomass (measured as chlorophyll a).
8 Ecology, Microbial

process occurs at rates of nmol cm3 year1, and enrichment cultures of the consortium of methanogen and sulfate-reducing
bacteria have a doubling time of 7 months, whereas when grown axenically, these organisms double in o 40 hours.
Principle 4. The results of microbial catalysis can have profound effects on the physical and chemical characteristics of the
macroenvironment. However, microbial activity is determined by the physical and chemical characteristics of the microenvironment.
There are many habitats in which spatial heterogeneities at the submillimeter scale produce a richer diversity of habitats and
niches than is apparent from measurement of bulk average concentrations of chemicals. In the open ocean, there is a spatial
mosaic of nanometer-sized organic colloids as well as sheets of organic compounds that may be up to 100 mm in size. Aggregation
of organic matter and organisms into millimeter-sized particles create a habitat for a rich and diverse community of phyto-
plankton, protists, and bacteria. In well-aerated soil, local deposition of labile organic matter will create a site of intense activity by
aerobic bacteria in which the rate of oxygen consumption exceeds its diffusion and an anoxic microsite in a soil particle can arise.
Strong concentration gradients in environments that result from physico-chemical processes, high densities of active microbes,
or (most commonly) the interaction between abiotic and biotic forces also produce strong effects at the microscale. The gastro-
intestinal tract of insects is a fermentation chamber with a high concentration of microbes; the fermentation products (organic
acids) are absorbed through the wall of the gut to fuel the nutrition of the animal. However, O2 can also diffuse from the trachea
across the gut wall, so that much of the gut volume is microoxic rather than fully anaerobic. Furthermore, H2 is a fermentation
product that arises in the anoxic lumen of the gut and is not used by the insect. It is an energy source for aerobic chemoautotrophic
bacteria, which have a niche where overlapping gradients of O2 and H2 occur in the lumen. In dense microbial mat systems,
diurnal changes in light irradiance drive changes in the rate of photosynthetic oxygen evolution that result in rapid temporal
changes over millimeter distances in the profiles of O2, H2S, and pH in the mats.
Principle 5. The microbial world exhibits much greater metabolic versatility than is found among macroorganisms. This
versatility is exhibited in several different ways:

• Essentially all reduced chemicals found on Earth (both organic and inorganic) can be used as an energy source by some
microbe if its oxidation can be coupled to reduction of a terminal electron acceptor in a thermodynamically favorable reaction.
Many biogeochemical cycles operate because some microbe can extract biologically useful energy from the reduced form of an
element (e.g., H2S) and transfer the electrons to a molecule (such as O2) that has a greater tendency to become reduced
(the terminal electron acceptor). The oxidized product (SO42) can then be reduced to H2S when it serves as a terminal electron
acceptor for a different physiological group of microbes.
For organic compounds, all naturally occurring organic compounds and most synthetic organic chemicals can be used as energy
sources by microbes. It is easier to list the recalcitrant exceptions – lignin (under anaerobic conditions), highly halogenated
organic compounds (under aerobic conditions), branched alkanes, and insoluble synthetic polymers (e.g., polyethylene).
• Individual organisms may be capable of metabolizing a broad range of different substrates.
From an ecological perspective, much of the organic matter that becomes available to chemoheterotrophic bacteria arises via cell
lysis from the death of other organisms (plants, animals, or other microbes). The enzymatic hydrolysis of cell macromolecules will
produce a rich mixture of various sugars, amino acids, and organic acids. Thus, it is not surprising that individual microbes may
possess transporters and enzymes that provide the potential for utilization of many components of these mixtures.
The study of biodegradative enzymes in bacteria has provided insights into some mechanisms whereby a broad substrate utilization
range may arise. In some cases, a key enzyme with a broad substrate range has evolved. A halidohydrolase from an Arthrobacter sp. is
active against more than 50 different halogenated organic compounds that differed in their halogen substituent and ranged from
aromatic to saturated or unsaturated alkanes. The evolution of biodegradative metabolic pathways has also made it facile for an
organism to utilize multiple organic substrates. There are a large diversity of substituted aromatic compounds that can be aerobically
catabolized by microbes. Catabolic pathways entail the action of enzymes that funnel different substrates to one of a few common
intermediates (usually dihydroxy benzene derivatives), which are then further catabolized by a common set of enzymes. Thus, the
growth substrate range can be augmented by acquisition of enzymes to process specific substrates, once it possesses a core aromatic
catabolism pathway.
• Complex microbial communities may be metabolically redundant and contain a number of different populations that can
metabolize the same substrate. This metabolic diversity increases resiliency to environmental perturbations, as extreme con-
ditions are less likely to kill all species performing a particular ecological function.
A critical issue in all areas of ecology is to understand the relationship between biodiversity and the functioning and stability of
ecosystems. Imagine an ecosystem that is devoid of microbes, and that different species are introduced one by one. Initially,
each new species is likely to add a new function to the ecosystem, because it can utilize substrates that the previous species
could not. At some point, however, new species are likely to be redundant – an existing species can already carry out that
function. The ecosystem possesses its maximum microbial functionality. However, higher biodiversity (with its associated
functional redundancy) is valuable if the ecosystem is exposed to changes in environmental conditions such as a change in pH
or addition of a toxic metal. If functional redundancy were low, then all species that can carry out a particular function might be
eliminated. At higher functional redundancy the probability that at least one species is tolerant of the new environmental
condition is higher and the ecosystem then retains this function and is perceived as functionally stable.
• Microbial populations are capable of rapid evolutionary responses to environmental perturbations; this is due to mechanisms
that generate relatively high mutation frequencies and a capacity for horizontal gene transfer (HGT) across large phylogenetic
boundaries.
 Ecology, Microbial 9

Evolution via differential reproduction under environmental selection pressures was the central premise of Darwin’s work. In
the case of plant and animal species, genetic variation (the raw material upon which selection acts) is generated within one species
via the shuffling of genes during sexual reproduction. In microbes, however, there are mechanisms of HGT that can cross broad
phylogenetic boundaries and generate completely novel gene combinations in much shorter time frames.
Humans have performed two great natural experiments that illustrate this point – the large-scale introductions of antibiotics
and xenobiotic organic compounds. Antibiotic-resistant forms of a pathogen have often arisen within 10–30 years after intro-
duction of a new antibiotic. The acquisition of multiple antibiotic resistance by clinical isolates is conferred by resistance transfer
factors (RTFs), broad host range plasmids that contain a set of transposons, each of which has a gene for resistance to a different
antibiotic. From where did these plasmids arise? A set of clinical isolates of Enterobacteriaceae that had been archived in the
preantibiotic era was found to contain plasmids, and their core genes for replication, maintenance, and conjugal transfer were
similar to those found in modern RTFs. However, they did not contain transposons with antibiotic resistance determinants. The
implication of these findings is that the use of antibiotics in clinical practice and agriculture results in a strong selection pressure
for cells with RTFs in which the appropriate transposon has integrated.
Why can microbial evolution proceed more quickly than in the case of macroorganisms? Microbial generation times can be
comparatively short, and each replication of genetic material will result in a small number of mutations. When one considers the
total population size of microbial cells in habitats (millions to billions per gram), there are ample opportunities for genetic
variation to arise.
The ‘evolution’ of specific taxa is particularly enhanced by bacterial mechanisms of gene transfer (the uptake of naked DNA
by a cell, or transfer of DNA into a recipient bacterium by direct contact with a donor bacterium or infection with a bacterial
virus) that may permit horizontal transfer of genetic material across broad phylogenetic barriers. Gene transfer frequencies are
directly proportional to cell density – crowded habitats such as animal gastrointestinal tracts, soil microcolonies, the plant
rhizosphere, biofilms, and marine snow probably represent important foci. In some instances, the acquisition of a single gene
(e.g., an antibiotic resistance gene or exotoxin genes within prophage of Corynebacterium diphtheriae or Clostrdium botulinum)
can substantively alter the organism’s ecological properties. In other instances, there are functional modules of multiple genes
that confer novel properties. Some examples include sets of genes for organic catabolism or pathogenicity islands that contain
virulence genes.

Principles in Population Ecology

Principle 6. The types and numbers of different microbes present in a habitat are a function of the diversity and amounts of
nutrient resources available; in ecology, this is termed bottom-up control. However, the abundance of a microbe will also be
modulated by its ability to withstand forces that eliminate it from the habitat. These loss forces may be biotic (due to the activity of
predators and parasites and termed top-down control), physical (flushing or scouring action), or physiological
(nutrient starvation).
The effect of nutrient resources upon microbial activity and growth will be discussed below in the context of physiological
ecology. Although much of the historical effort in microbial ecology has been directed to the study of bottom-up control, there
have been significant studies since the 1980s on the effects that biotic loss factors may have on structuring microbial communities.
The recognition of the role played by eukaryotic protists in biogeochemical cycling in the ocean was important in this regard.
Bacteria have been found to utilize up to 50% of the carbon fixed by photosynthesis in the ocean. The ‘microbial loop’ (Fig. 5)
provides a means for these resources to be reintroduced into the traditional food chain (via zooplankton to fish). Bacterial
numbers are controlled by nanoflagellates and to a lesser degree small ciliated protists, which in turn are preyed upon by
microzooplankton. As these predators are of similar size to phytoplankton, they can be consumed by zooplankton and the
resources reenter the planktonic food chain. Detailed investigation of other habitats has shown that bacteriovorous protists are
important in a range of ecosystems, including soil.

Fig. 5 The microbial loop in aquatic food chains. Filter feeding by protists upon bacteria can result in the return of primary production back to
the classical food chain, as the protists are preyed upon by zooplankton. POM, particulate organic matter; DOM, dissolved organic matter.
10 Ecology, Microbial

There is a vigorous debate between those who contend that bottom-up mechanisms control microbial biomass levels and those
who emphasize top-down control. The truth likely lies in the middle, and may differ depending upon season and ecosystem
productivity. Both empirical and theoretical evidence suggest that bacterial numbers may be more tightly controlled by protistan
predation in nutrient-poor (oligotrophic) ecosystems, whereas in nutrient-rich habitats competition for nutrients has primacy.
Intuition might lead one to the opposite inference, but eutrophic habitats will have higher population levels of higher-level
predators (zooplankton and fish) that keep tighter control on the abundance of bacteriovorous protists. As a result, predation
pressure on bacteria is reduced and higher bacterial numbers result in more stringent nutrient limitation.
Protistan grazing not only can control bacterial numbers, it can also shape bacterial community composition. Protists employ
either filter or interception feeding. As a result, bacterial cells with diameters of 1–3 mm are most susceptible to predation. And,
therefore, aquatic protists are very effective predators on rapidly growing single-celled bacteria. Thus, there will be strong selective
pressure for antipredator strategies in bacteria that result in smaller or larger particles. Strategies that yield larger particles include
morphological changes such as filament formation or production of exopolymers to produce cell aggregates. Ultramicrobacteria
(those with a cell diameter of 0.3 mm or less) can be isolated from natural marine samples; these organisms are sized below the
optimal feeding range of protists. Chemical defenses have also been found – the purple pigment of Chromobacterium violaceum has
been found to induce apoptosis in flagellates when a bacterial cell is ingested.
Over the past 25 years, aquatic ecologists have come to recognize that viruses are also significant causes of microbial mortality.
There are approximately 108 virus particles per milliliter in productive, coastal marine waters. There is significant genetic diversity
among these viral particles, and within that diversity are viruses that can infect algae, bacteria, or protists. It has been technically
challenging to accurately measure the proportion of microbes that are lysed by viruses; current estimates range from 10% to 50%
of the bacterial population per day and the impact of viruses as a loss factor for bacterial production may equal that of
bacteriovorous protists in marine systems.
Principle 7. Microbial populations and communities often exhibit much larger dynamics in biomass, composition, and activity
than do plant and animal populations.
These complex, nonequilibrium dynamics are thought to be primarily driven by the temporal frequency of changes in
important environmental factors (and hence, changes in selective environmental forces). An obvious example is a tidal salt marsh,
but there are periodicities that exist at shorter and longer frequencies to which microbes can respond. Temporal changes can occur
over very different spatial scales, from kilometers in coastal marine upwellings to millimeters in laminated microbial mats. In
addition to these environmentally driven changes, there are also internal biological cycles in ecosystems that alter selective forces.
Predator–prey cycles are perhaps the best known, but the identification of cross-species molecular signaling in the laboratory
suggests that population dynamics may have a larger biological component than we have appreciated.
Microbes respond more dynamically to these changes than macroorganisms as a consequence of their biological properties: a
rapid pace of physiological adaptation to changing environmental conditions, a broad variety of positive and negative interactions
between different microbes and the capacity for rapid genetic change. An important consequence of principles 6 and 7 is that the
observed temporal changes in microbial populations (whether termed ‘community dynamics’, ‘successions’, or ‘oscillations’) are
not solely determined by environmental or biological factors, but rather by an intricate interaction among these factors.

Principles in Physiological Ecology

Microbial physiological ecology is a field in which the molecular details of gene regulation can be evaluated in the context of
ecological effects under the selective pressures experienced by microbes in nature.
Principle 8. The amount of one essential limiting resource is most likely to determine the amount of microbial biomass in an
ecosystem.
A nutrient resource that is taken up may be (1) conserved in functional or structural molecules (i.e., become biomass), (2)
utilized and excreted for bioenergetic purposes (energy source and terminal electron acceptor), or (3) function both as a com-
ponent of biomass and in bioenergetics. All organisms must acquire an energy source (light of the appropriate wavelengths for
phototrophs and a reduced chemical for chemotrophs) and the appropriate molecular form of atoms that comprise biomass. The
most important atoms are C, N, P, S, and several metals (Fe is often the most quantitatively important) that serve as enzyme
cofactors. Chemotrophic bacteria also require a terminal electron acceptor, a molecule to which electrons are ultimately donated
after oxidation of the energy source. Autotrophs require reducing power to reduce CO2 ; in photoautotrophs an external electron
donor is used for this purpose via photosynthesis whereas chemoautotrophs may use their reduced energy source directly or
generate reductant via reverse electron transport.
The molecules that comprise this elementary set of requirements can be called ‘complementary substrates’ in that all are
necessary, but none by themselves are sufficient for growth. However, within each requirement, there may be substitutable forms.
All microbes use NH4 þ as an N source, a number can use NO3, and some fix N2. There are hundreds of different organic
substrates in the environment; if a specific microbe has uptake systems and enzymes to catabolize them, these represent
substitutable substrates.
The limiting nutrient is not necessarily the one present in the lowest absolute concentration in the habitat. Rather, it is the
resource in scarcest supply relative to its requirement for biomass synthesis (Liebig’s law of the minimum). It is important to
recognize that cellular composition is not immutable. Despite the general biochemical similarity of all organisms, some taxa may
 Ecology, Microbial 11

have unique requirements. For example, diatoms require silica for their extracellular frustules whereas other groups of phyto-
plankton do not have this requirement. In addition, physiological conditions may substantially affect the elemental stoichiometry
of biomass. Iron is required in a variety of electron transfer proteins such as cytochromes and nonheme iron proteins; it usually
comprises 0.2% or less of a bacterium’s dry weight. However, in N-limited environments where N2-fixing bacteria would have a
selective advantage, the synthesis of Fe-rich nitrogenase and ferredoxin would increase cellular iron content by at least fivefold.
Principle 9. Growth of microbes on the limiting nutrient (or a set of substitutable nutrients) will depress its concentration to a
value that limits microbial growth rate. The nutrient flux rate in an ecosystem may become physically limited by hydrodynamics or
mass transport.
The ‘limiting nutrient’ in a habitat has two distinguishable effects on microbial growth. It limits the total amount of biomass. If
microbes are inoculated into a sterile environment, they will grow at their maximum rate and at some point the total demand of
biomass for nutrient uptake exceeds its supply rate, and the concentration of limiting nutrient decreases. At some low con-
centration (in the nmol l1 to low mmol l1 range), transport rate into the cell is determined by this low extracellular substrate
concentration and a reduced intracellular flux produces a reduced growth rate.
Principle 10. Competition among microbes for low concentrations of limiting substrate and temporal variability in nutrient
availability is a strong selective force in natural habitats. A variety of physiological and morphological adaptations can have
adaptive value under specific conditions.

1. The number, diversity, and affinity of transporters in the cell membrane: Nutrient concentrations in the nmol to low mmol l1
range are likely to constrain the growth rate. The organisms that can maximize nutrient flux into the cytoplasm will have the
highest growth rate and therefore a competitive advantage. In the case of chemoheterotrophic bacteria in the open ocean,
organic substrates are likely to be the limiting nutrient. The dissolved organic C level in seawater totals B1 mg C l1, but this
represents nmol l1 concentrations of a wide variety of specific substrates derived from the lysis of phytoplankton cells and the
hydrolysis of their macromolecules. One physiological strategy would be to express genes for high-affinity transporters, because
at these low external substrate concentrations, the specific affinity (the initial slope of the uptake rate vs. substrate concentration
function) provides a competitive advantage. Furthermore, derepression of gene expression to increase the number of
transporters inserted in the membrane will enhance substrate flux via the law of mass action. The maximum rate of transport
(Vmax) normalized to a per cell basis is a function of the number of transporters per cell. Increases in Vmax not only provide a
higher transport rates at saturating substrate concentrations, but also increase flux rates at rate-limiting concentrations (Fig. 6).
When there are approximately 1500 transporters of one specific type in the membrane, they compete with each other for
substrate molecules, and further investment in additional transporters of this type is not productive. However, only 2% of the
membrane surface is covered with transporters; synthesizing transporters with many different specificities will maximize the total
flux of limiting organic C into the cell, if other organic substrates are available. Thus, organic C-limited chemoheterotrophs are
predicted to express the capacity for the transport and catabolism of multiple organic substrates. They need only generate a fraction
of their total growth rate from the transport flux of each individual organic compound.

2. Regulation of the capacity for substrate uptake versus substrate assimilation: The capacity for nutrient uptake is maximized by
increasing the number, diversity, and specific affinity of transporters. However, in a competitive environment this will push ambient
substrate concentrations to values where the growth rate is far below maximal. The optimal use of resources under these conditions
will regulate decreased amounts of the macromolecular machinery for biosynthesis and assimilation. The most obvious example of
this is the direct relationship between growth rate and cellular ribosome content. Ribosomes are energetically expensive to produce; a
cellular content that exceeds the required rate of protein synthesis is energetically inefficient.

Fig. 6 The effect of increasing the number of membrane-bound permeases per cell, at both low, transport-limiting external substrate
concentrations and saturating concentrations. Calculations were made employing the Michaelis–Menten equation at Ks of 10 mmol l1, and
different Vmax values. Vmax is proportional to the number of permeases per cell.
12 Ecology, Microbial

3. Cell size: When bacteria are grown in a chemostat at a variety of growth rates, slow-growing cells have a smaller size and
volume than fast-growing populations. The direct physical reason for this may be the differences in ribosome content
mentioned above. A tenfold increase in cellular ribosome content can only be accommodated by increasing the volume of a
cell. Direct microscopic examination of bacteria in aquatic environments indicates that many cells are quite small, with
diameters of 0.5 mm. Size-selective predation by protists may play a role here, but the low growth rates typically found in
planktonic aquatic ecosystems (doubling times are often measured in the range of 16–50 h) also lead to small cells. The
physical nature of small cells has another beneficial effect for nutrient-limited habitats – the surface-to-volume ratio increases as
cell diameter decreases. Membrane-bound surface area is the location of substrate transporters, whereas the cytoplasmic volume
contains soluble biosynthetic machinery. Thus, decreases in cell size found when nutrient-limited growth is slow have the
consequence of increasing the ratio of transporters-to-assimilation machinery.
4. Storage polymers: The physiological responses to nutrient limitation in nature include derepression of nutrient transport
systems and downregulation of biosynthetic pathways. However, a microbe may be exposed to pulses of higher nutrient
concentration either on a regular diurnal cycle or at irregular intervals (e.g., by encountering particulate organic matter in an
aquatic ecosystem). At this physiological state, transport of the limiting nutrient occurs at a much faster rate than can be
assimilated into proteins and nucleic acids. The solution to this mismatch is intracellular storage as a polymer that can be
assimilated later when external substrate concentration is again limiting.

Storage polymers of C and energy include glycogen, starch, and polyhydroxyalkanoates. P is stored as polyphosphate by a
number of microbes. N is the macronutrient for which there is no storage polymer, with the exception of cyanophycin produced in
many cyanobacteria and a few heterotrophic bacteria. The polymer consists of an aspartic acid backbone to which arginine
residues are linked.
The synthesis and degradation of glycogen by the cyanobacterium Oscillatoria aghardii during a diurnal light–dark cycle
illustrates the interplay between resource acquisition, storage, and cell growth. During the light period, fixed CO2 is used for
immediate growth (synthesis of proteins and nucleic acids) and stored as glycogen. Of particular note is that growth rate
(measured as rate of protein synthesis) during the light period is lower than in cells cultured in continuous light. However, during
the dark period the intracellularly stored glycogen is used as both a carbon and an energy source – the rate of protein synthesis in
the dark is very similar to that measured in the light. That is, the microbes grow chemoheterotrophically in the dark on stored
glycogen. This physiological strategy produces a higher daily growth rate than if the cyanobacteria had a growth rate of mmax in the
light and 0 in the dark.

Strategic Approaches to the Study of Microbial Ecology

An examination of articles published in microbial ecology journals will illustrate not only the breadth of habitats that are studied
by microbial ecologists, but also a variety of technical approaches. Some of the major categories are described below.

Descriptive Studies of the Distribution and Activity of Microbes in a Natural Habitat

‘Descriptive studies’ has become a pejorative term in science, but this type of analysis is a necessary first step when initiating a study
on a new habitat. In addition, field studies have an overhead not found in well-controlled laboratory studies – the collection of
routine data on physical, chemical, and biological characteristics of the system. It is essential to monitor the natural variation that
lies behind any hypothesis-driven field experiments. These routine data can be mined using statistical techniques to uncover
relationships between environmental factors and biological activities. These relationships suggest hypotheses that can be
experimentally tested.
The most common types of descriptive data published in ecological studies are lists of microbial community members
identified from sequences of 16S rRNA gene amplicons or metagenomic analyses. About 23% of the 22,000 papers published on
microbial ecology over the past 10 years have included at least one of these data types. Although knowledge of community
composition on its own is useful, it is far more impactful to conduct deeper analysis of datasets comprising hundreds of samples.
By employing statistical frameworks, changes in microbial community composition in space or time can be used to infer the
relative importance of ecological forces (dispersal, selection, and drift) that shape microbial composition. Furthermore, when
related to field-scale measurements of physical and chemical variations, the effects of environmental variable variation upon the
intensity of those ecological forces can be mapped.

Analysis of In Situ Rates of Growth or Biogeochemical Activity

These analyses can rarely be accomplished under true in situ conditions, as they entail addition of chemical tracers or measurement of
changes in biomass. Thus, natural samples are collected and confined in containers, but incubated under the physical conditions found
in situ. The removal of a sample from the open conditions of the natural system to a confined bottle will lead to changes with time.
Therefore, it is advantageous to keep the incubation time as short as possible. The use of radiotracers (i.e., the addition of a radioactive
molecule in an amount that is small enough to have a negligible impact on in situ chemical concentration) has been invaluable in rate
 Ecology, Microbial 13

measurements of biogeochemical processes. Bacterial growth rates have been estimated by measuring the rate at which radioisotopic
forms of biosynthetic precursors (such as thymidine or leucine) are incorporated into cellular macromolecules. These calculations
require several assumptions, and give an ‘average’ growth rate for a heterogeneous population of microbes. These results generally
suggest that the doubling times of aquatic bacterioplankton are in the range of 1–2 days, much longer than what is achievable with pure
cultures in the laboratory. In situ growth rate is a fundamental parameter of interest in ecological studies, but it remains a difficult one to
accurately and conveniently measure for microbial populations in nature.

Experimental Manipulation of Natural Samples Under Simulated Conditions (Microcosms)

The intent of microcosm experiments is generally not to accurately measure in situ activity rates, but rather to perform experimental
manipulations of natural microbial communities to test hypotheses regarding the effects of environmental perturbations upon
microbial community activity or composition. Incubation times are likely to be much longer than when measuring in situ activity
rates; therefore, caution in interpretation is warranted as confinement of the sample can introduce major effects.
An example study examined the population structure and activity of a microbial community in soils that were cocontaminated
with lead, chromium, and aromatic hydrocarbons. It was difficult to separate the effects of different contaminants by direct
analysis of spatial differences in community composition and environmental contaminants. Therefore, microcosm studies were set
up in which soil samples from both contaminated and uncontaminated sites were challenged with different terminal electron
acceptors and a graded series of heavy metal concentrations, to analyze the effects upon microbial activity (the mineralization of
organic carbon) and community composition. The experiments illustrated that if one wanted to detect changes in community
composition, it was important to add an energy source (organic substrate) that could drive significant changes in composition. Of
course, this energy source will itself represent a strong selective force in structuring the microbial community. However, by
comparison of sample treatments, it was possible to demonstrate that whereas resource factors such as the organic energy source
and terminal electron acceptor were a major force in structuring species richness, the effects were modulated by toxic factors such
as heavy metals. In addition, the factorial design of the experiments (in which a series of combinations of energy sources, terminal
electron acceptors, and heavy metal concentrations were tested) showed that microbial activities carried out by a diverse group of
functionally redundant community members were more resistant to high toxic metal inputs than activities for which a more
restricted diversity of microbes was likely to be present.

Model Laboratory Systems

A distinction between microcosm experiments and model laboratory systems is that the former are initiated with many of the
physical and chemical characteristics as well as the taxon richness of the original habitat whereas the latter is often a highly
simplified version of nature. In many cases, it may be the study of a pure culture in the laboratory. This can be controversial in
microbial ecology, because the laboratory system will lack a large number of features found in nature. However, ecophysiological
studies in the laboratory provide the means to rigorously ascertain the physiological responses made by a microbe to environ-
mental changes. These can be approached with a degree of experimental control and replication that is not possible in the habitat.
If the study remains housed entirely in the laboratory, it probably ceases to be an ecological one. However, there are advantages to
integration of field and laboratory studies, and recursion between real-world complexity and the experimental control that can be
applied in the laboratory.
Ecological studies on planktonic cyanobacteria have provided an example of the value of this type of recursion. In eutrophic
lakes, cyanobacteria proliferate to high cell densities (blooms) in which 490% of the biomass in the sample consists of one
morphological type of cyanobacterium. Under these unique conditions, one can measure physiological properties in whole
samples (photosynthetic rate, allocation of fixed C to different macromolecules, rates of macronutrient uptake) and reasonably
ascribe these properties to the dominant biomass component. From these measurements on field populations, hypotheses arise
regarding the regulatory responses made by the organism to limiting environmental factors. These hypotheses are testable in the
laboratory by experiments with cyanobacterial isolates cultivated in chemostats under control of growth rate and stringency of
nutrient limitation. As deeper insights are gained into the details of ecophysiological responses from laboratory experiments, it is
important to revisit the field and ascertain that those details actually operate under field conditions. This iteration between
laboratory and field is straightforward when the natural population has a high relative abundance of the organism of interest. A
remaining challenge for microbial ecology is to develop technologies that allow application of this iterative approach to com-
munity members present at low relative abundance. This requires the capacity to rapidly detect specific taxa and make physio-
logical measurements on single cells.

Future of Microbial Ecology

The Microbiome
There has been an explosion in the past ten years of work on the ‘microbiome,’ the microbial community in a specific habitat. As
often occurs in science, explosions are driven by technology developments – in this case, fast and cheap high-throughput DNA
14 Ecology, Microbial

sequencing coupled with bioinformatics algorithms, databases, and methods of data analysis. Currently, the most common
approaches after isolating DNA directly from environmental samples are PCR amplicon sequencing of a portion of the 16S rRNA
gene and shotgun metagenomic sequencing. These methods have produced tremendous insights into the diversity of microbes and
their functional genes, and will continue to be a major component of microbial ecology research in the future. However, these
molecular approaches have not yet reached the aspirations held for them to unlock deeper insights into microbial ecology. The
issues include:

1. DNA sequence information provides insight into metabolic potential, but not actual activity. For example, a high proportion of
nif genes in a metagenome does not indicate that nitrogen fixation actually occurs in the habitat. The dominant biota may have
been selected for other traits, but happen to also possess nif genes. Advances in the practicality of metatranscriptome and
metaproteome analyses would address this gap.
2. Most studies do not conduct adequate sample replication in site, space, or time to derive statistically rigorous conclusions.
3. Hypothesis testing typically involves multivariate statistical analysis of molecular sequence and environmental data. However,
these approaches can only indicate correlation but never prove causation.
4. The technical choices for DNA isolation method, primers, and sequencing technology all introduce biases. Hence, comparisons
across laboratories and historical studies is difficult. The Earth Microbiome Project is attempting to introduce standards for
these issues.
5. Reference databases are inadequate. There is a bias in databases (both for taxa and genes) towards certain relatively well-
understood phyla (for example, the Proteobacteria). As a result:
a. The commonly used 97% sequence similarity threshold for delineating “taxa” from 16S rRNA sequences probably blurs
or does not capture the breadth of microbial ecological heterogeneity.
b. In terms of inferring potential microbial physiology from metagenomes, the inadequacies of reference databases mean
(a) some annotations are wrong, (b) gene calls are conservative, hence 30–50% of genes found are of undetermined
function, and many others identify a protein family rather than a specific enzymatic activity, (c) only environmental genes
with 440% identity to a homologous database sequence are very likely to share functional similarity.

Microbiology at the Microscale

It is a central principle that the direct interaction of microbes with their environment occurs at a spatial scale of micrometers. Thus,
analysis of the authentic ecology of microbes is enhanced by technologies that increase the spatial resolution of experimental
measurements. This applies not only to analysis of the physical and chemical environment around a cell, but also to measuring the
physiological and genetic properties of a set of individual microbial cells. In addition, analysis of a set of individual microbial cells
can provide the basis to understand not only physiological responses at the microscale, but also the distribution of properties
among a population of cells within a natural habitat.
Future possibilities include the continued development of microfluidic devices that can be designed to impose gradients of
environmental factors or simulate natural microhabitats (e.g., soil pore network or gastrointestinal villi). Optical sensors for
important metabolic reactant (such as O2 and CO2) can be incorporated into such devices.
To achieve microscale microbiology, the development and application of high-resolution imaging approaches that couple
biochemical analyses to micron scale interactions of microbes with each other and their environment is critical. There have been
impressive advances using a variety of approaches in analytical chemistry. These include label-free techniques such as Raman and
FTIR microspectroscopy which have been applied to the analysis of microbial cells, and nuclear magnetic resonance that can be
used to image biofilms and also provide information on chemical gradients within them. New fluorescence probes and optical
imaging systems provide the capacity to interrogate physiological state, gene expression, and phylogenetic identity of individual
cells in a natural sample.
Radioisotopes have been a powerful technology in advancing microbial ecology; however, they also have practical limitations
for performing in situ experiments and handling waste. Stable isotope methodologies have the potential to revolutionize the field
of microbial ecology. Sophisticated improvements in nuclear magnetic resonance, mass spectrometry, and Raman spectroscopy are
approaching the point that allow the kinetics of metabolic function in microbial communities to be determined with great
sensitivity. In contrast to radioisotope-based techniques, stable isotopes provide a means to specifically isolate cellular macro-
molecules that have incorporated the label and therefore have a unique signature. There remains great scope for technical
refinements and advancements in the use of stable isotopes for metabolic and molecular analyses of microbial communities; these
future methodological innovations will provide the means to mechanistically understand in situ function of microbial commu-
nities in a quantitative way. With technology developments regarding sensitivity and spatial resolution, this has the potential to
provide a direct link between measurements of community function (metabolism) and expression of functional genes
(via labeling and analysis of either mRNA or protein). An example of this has been the application of nanoSIMS (secondary ion
mass spectrometry) in microbial ecology, which makes possible the determination of chemical and stable isotope composition at
the submicron level.
Single-cell microbial genomics is now a reality, but remains tedious due to current sequencing technologies. If accurate single
molecule DNA sequencing comes to fruition, imagine that 300 individual microbes from a habitat are physically separated, and
each are subjected to single-molecule genome sequencing. One has in hand a community metagenome in which environmental
 Ecology, Microbial 15

sequences are inherently placed in their genomic context, and information on the relative abundance of different genomes
(and genetic microdiversity within a specific clade) is directly available.

Theory in Microbial Ecology

Technology is not enough. New technologies will generate a blizzard of data at an ever-increasing rate, but understanding can only
come from a synthesis of that information. Data-rich technologies carry the risk that they drive science to idiosyncratic pathways
that do not yield deep understanding of fundamental principles. The antidote is to ground studies to sound theory, a synthesis of
how we imagine microbial populations and communities operate. In comparison to the ecology of plants and animals, microbial
systems are much more experimentally tractable, which provides the opportunity to test theoretical syntheses that are either
unique to microbial ecology or derived from the rich body of macroorganismal ecological theory.
There are several means to articulate a theoretical synthesis; a simulation model is one of them. This approach has the virtue of
forcing a qualitative, descriptive discipline to become a quantitative, predictive one. The challenges are great, particularly at the
community level, because functional redundancy will make precise, unambiguous prediction of community composition
uncertain. However, robust model predictions may hold at the cellular, population, and ecosystem levels and lead to deeper
insights into principles of microbial ecology than presented here. The greatest value of a simulation modeling approach is an
iterative interaction with experimentation. The test of a simple model against ecological reality illuminates its weaknesses, and
aspects in which greater depth of detail is necessary. Sensitivity analyses of a refined model suggest the most insightful experiments
to carry out in the field, which lead to another round of model refinement.

Does Cultivation Have a Place?

For more than 30 years, the conventional wisdom in microbial ecology is that less than 1% of bacteria in nature can be cultivated.
That observation in fact was a driving force for the development of cultivation-independent analyses of nucleic acids that are now
the driving force in the field. Does there remain a role for cultivation?
There have been important innovations that can in many cases raise the proportion of cultivatable microbes from a habitat to
10%–25%. Culture chambers that mimic natural environments (for example, by using microfluidic devices) have been successful.
Co-culturing has also been effective. Techniques that employ substrate concentrations at in situ levels or the addition of substances
such as signaling molecules or resuscitation factors have been successful in cultivating relatively high proportions of the
population. However, it is true that cultivation remains a time-consuming undertaking that is resists automation.
But cultivation still plays two important roles in microbial ecology, that complement molecular analyses of the microbiome.
First, it provides the raw material to improve the genome reference databases that metagenomics relies upon. Secondly, the future
entails understanding the functional potential of individual microbial ecotypes (metabolic versatility) as well as functional
redundancy in microbial communities, and here molecular approaches have fundamental weaknesses. In particular, metagenomic
approaches can uncover novel gene sequences or identify familiar ones, but in the end only allow the formulation of hypotheses
that still must be subjected to biochemical analysis for rigorous proof. In contrast, propagation of an isolate under controlled
conditions permits application of a broad array of tools to investigate its specific ecophysiology and genetics.

Further Reading

Azam, F., Malfatti, F., 2007. Microbial structuring of marine ecosystems. Nature Reviews Microbiology 5, 782–791.
Battin, T.J., Sloan, W.T., Kjelleberg, S., et al., 2007. Microbial landscapes: New paths to biofilm research. Nature Reviews Microbiology 5, 76–81.
Brune, A., Frenzel, P., Cypionka, H., 2000. Life at the oxic-anoxic interface: Microbial activities and adaptations. FEMS Microbiology Reviews 24, 691–710.
Button, D.K., 1998. Nutrient uptake by microorganisms according to kinetic parameters from theory as related to cytoarchitecture. Microbiology and Molecular Biology Reviews
62, 636–645.
Fenchel, T., King, G.M., Blackburn, H., 1998. Bacterial Biogeochemistry: The Ecophysiology of Mineral Cycling. San Diego: Academic Press.
Madsen, E.L., 2005. Identifying microorganisms responsible for ecologically significant biogeochemical processes. Nature Reviews Microbiology 3, 439–446.
O’Donnell, A.G., Young, I.M., Rushton, S.P., Shirley, M.D., Crawford, J.W., 2007. Visualization, modelling and prediction in soil microbiology. Nature Reviews Microbiology 5,
689–699.
Prosser, J.I., Bohannan, B.J.M., Curtis, T.P., et al., 2007. Essay – The role of ecological theory in microbial ecology. Nature Reviews Microbiology 5, 384–392.
Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: The Biology of Elements From Molecules to the Biosphere. Princeton, New Jersey: Princeton University Press.