Content uploaded by Patricia Geesink
Author content
All content in this area was uploaded by Patricia Geesink on Dec 11, 2020
Content may be subject to copyright.
Advances in genome- sequencing technologies and
sophisticated metagenomics and phylogenetic meth-
ods have contributed to drastically change our views
on thediversity of microbial life, including the very
shape of the tree of life1,2. Yet the marked expansion of
genomic data, which has led to an improved understand-
ing of archaeal and bacterial diversity, is contrasted by
our inability to culture representatives for many of the
novel lineages. Consequently, most of what we currently
know about archaea and bacteria is either derived froma
minority of well- studied cultured lineages or from
reconstructed genomes belonging to uncultured line-
ages. Although this period of rapid, genome- driven dis-
covery has provided numerous important new insights
into microbial life on our planet, it is essential to isolate
and culture species from these uncultured lineages to
test genome- based predictions about their cell biology
and physiology and to properly understand their eco-
logical roles. Such a need is emphasized by examples of
completely novel enzymatic reactions and pathways that
have been discovered through experimental testing of
microbial enrichments or cultures3–5, and some of these
pathways were undetectable by genomic methods alone4.
Microbial cultivation can be used to generate
pure cultures, which provide a continuous supply of cells
from the same species or strain. Such cultures can be
used to investigate microbial traits in experiments per-
formed in replicates, which improves reproducibility and
statistical confidence. Without pure cultures (or, in some
cases, highly enriched co- cultures that contain a small
number of species that depend on each other to grow)
it is difficult to accurately determine microbial features
such as growth characteristics, metabolism, physiology
and cell biology for a single organism. These features
are also difficult to infer from genome sequences alone
because genomic data provide no indication of which
genes are functionally expressed, and therefore no indi-
cation of how the active proteome adapts under certain
conditions. Although metatranscriptomics and meta-
proteomics can provide some insights, the data that
these approaches generate are still difficult to interpret
without fundamental knowledge of the underlying phys-
iology. Thus, without cultivation, many questions about
the role of organisms in their natural settings remain
unanswered.
To improve our understanding of the uncultured
archaeal and bacterial diversity, it is essential to increase
our capacity for bringing microorganisms from the envi-
ronment into culture6–9. To achieve this task in a more
efficient manner, prioritizing cultivation efforts for
microorganisms that are thought to be the most inter-
esting (for example, because they represent a poorly
characterized group) or are most likely to improve
our understanding of a particular process (TABLE1) is
ultimately required10.
Traditional microbiological methods (BOX1) are
hugely important and indispensable assets for culti-
vation, and are continually used to successfully iso-
late numerous microorganisms of interest. However,
these approaches often require substantial amounts of
time and patience to succeed, as well as painstaking
and meticulous testing of media combinations and of
different physicochemical conditions. To circumvent
or at least minimize these potential limitations, inno-
vative new technologies (some of which incorporate
and expand on classical methods) have broadened the
Enrichments
Assemblages of several
strains that evolve from a
taxonomically diverse inoculum
in response to controlled
environmental selection
pressures (such as substrates
or temperature).
Pure cultures
Cultures containing cells
belonging to the same strain,
ideally originating from a single
cell or colony, that have
minimal genetic variation
between them. Also often
called axenic cultures.
Co- cultures
Defined assemblages of two or
more strains, often artificially
introduced and grown together
in the laboratory, which may
establish interspecies metabolic
relationships with one another.
Innovations to culturing the uncultured
microbial majority
WilliamH.Lewis, GuillaumeTahon , PatriciaGeesink , DianaZ.Sousa and
ThijsJ.G.Ettema
✉
Abstract | Despite the surge of microbial genome data, experimental testing is important to
confirm inferences about the cell biology, ecological roles and evolution of microorganisms.
As the majority of archaeal and bacterial diversity remains uncultured and poorly characterized,
culturing is apriority. The growing interest in and need for efficient cultivation strategies has
led to many rapid methodological and technological advances. In this Review, we discuss
common barriers that can hamper the isolation and culturing of novel microorganisms and
review emerging, innovative methods for targeted or high- throughput cultivation. We also
highlight recent examples of successful cultivation of novel archaea and bacteria, and suggest
key microorganisms for future cultivation attempts.
Laboratory of Microbiology,
Wageningen University and
Research, Wageningen,
The Netherlands.
✉e- mail: thijs.ettema@wur.nl
https://doi.org/10.1038/
s41579-020-00458-8
REVIEwS
Nature reviews
|
Microbiology
Table 1 | Key targets for cultivation
Target
microorganism
or groupa
Common
environments
Superphylum or
phylum
Reasons they are of interest for cultivation
Archaea
Anaerobic
methanotroph
clades 1, 2 and 3
Sediments Euryarchaeota They function as an important sink for the greenhouse gas methane, which they
metabolize as it seeps out from methane reservoirs beneath marine sediments.
They therefore help to limit the amount of methane that is released into the
atmosphere and are the only known organisms capable of oxidizing methane
under anaerobic conditions127.
Bathyarchaeota Sediments TACK archaea They are a group of globally widespread metabolic generalists that are abundant
in anoxic sediments. They contain some of the few known putative methanogen
lineages from outside the Euryarchaeota41.
Verstraetearchaeota Sediments TACK archaea Some of the few known putative methanogens from outside the Euryarchaeota
belong to this phylum128.
Candidate phyla
Heimdallarchaeota,
Helarchaeota,
Lokiarchaeota,
Odinaracheota and
Thorarchaeota
Marine
sediments and
hydrothermal
vents
Candidate
superphylum Asgard
archaea
These archaea belonging to the Asgard superphylum are important for
understanding the origin of eukaryotes. The Heimdallarchaeota are currently
the best- supported sister linage of eukaryotes, and are therefore the most
important target for cultivation. Some lineages are also abundant in some marine
sediments124.
DPANN archaea Assorted DPANN archaea They are a major archaeal group, currently thought to consist of at least 12
different phyla, with 6 cultured representatives across the entire group. They
typically have small cell and genome sizes, limited metabolic capabilities and are
likely to be symbionts or parasites of other microorganisms129.
Marine Group II, III
and IV archaea
Marine Euryarchaeota Marine Group II are abundant in some marine environments and are thought to be
important for the degradation of organic carbon130. Marine Groups III and IV are
abundant and widespread in some marine environments, and there are currently
no cultured representatives for any of these clades131.
Water column B
Thaumarchaeota
Marine Thaumarchaeota They have a key role in biogeochemistry by participating in carbon and nitrogen
cycling in the deeper layers of oceans132.
Bacteria
Acidobacteria Soil Acidobacteria They are a widespread and abundant phylum of versatile heterotrophs, thought to
have a major impact on the ecology of some terrestrial environments133.
Candidate phylum
Rokubacteria
Soil Candidate phylum
Rokubacteria
They are a novel phylum with unusually small cell sizes but large genomes and are
widespread in terrestrial ecosystems134.
Candidatus
Actinomarinidae
Marine Actinobacteria (OM1) A class with no cultured representatives in the Actinobacteria (which otherwise
have numerous cultured representatives (FIG.1)). They have streamlined genomes,
ultra- small cell sizes and are putative photoheterotrophs as their genomes encode
genes for rhodopsins135.
Candidatus
Atribacteria
Sediments Candidate phylum
Atribacteria (OP9/JS1)
They are globally distributed, and in some environments are abundant, and contain
species that are thought to be anaerobic hydrocarbon degraders136 as well as some
that are thought to be syntrophic propionate oxidizers137.
Candidatus
Dormibacteraeota
and Candidatus
Eremiobacteraeota
Soil Candidate phylum
AD3 and Candidate
phylum WPS-2,
respectively
These novel phyla contain species that are thought to survive on the consumption
of trace atmospheric gases. Their cultivation could provide wider insight into the
growth strategies used by bacteria that are abundant in oligotrophic soils138.
Candidatus
Marinimicrobia
Marine Candidate phylum
marine group A
They are an abundant and highly diverse group, participating in sulfur and
nitrogen cycles, driving the biogeochemistry of oceans, and might also function
as a potential sink for the greenhouse gas nitrous oxide139.
Candidatus
Poribacteria
Marine Candidate phylum
Poribacteria
They are often dominant and widespread members of microbial communities
associated with marine sponges139.
‘Candidatus
Udaeobacter
copiosus’
Soil Verrucomicrobia They are metabolically efficient heterotrophs with unusually small genomes, which
are widespread and abundant in many soils140.
Dehalogenating
bacteria
Assorted Chloroflexi, Firmicutes
and others
Some of these bacteria have been shown to respire anthropogenic chemicals that
are common environmental contaminants, suggesting they could be useful for
bioremediation141.
CL500-11 Aquatic Chloroflexi Members of this clade are abundant globally in the low- temperature layers of deep
freshwater lakes142.
SAR202 Marine Chloroflexi They are abundant in mesopelagic and bathypelagic marine layers, where they are
thought to have major roles in sulfur cycles143.
www.nature.com/nrmicro
Reviews
toolkit for microbial isolation and the efficient determi-
nation of suitable culture conditions. Although many
of these technologies remain in their infancy, with
wide- ranging applicability having not yet been demon-
strated for diverse species and environments, other
technologies are already used widely by researchers and
are starting to have a positive impact on microbiologi-
cal research. In this Review, we revisit the capabilities
and limitations of traditional isolation and cultivation
methods, and provide an overview and discussion of
more recent innovative technologies that have potential
to improve our ability to isolate, culture and charac-
terize microorganisms from poorly studied groups.
We also highlight recent successes in culturing elusive
microorganisms and present a list of some examples
of microorganismsto prioritize in future cultivation
attempts.
Recent successes in cultivation
In recent years there have been a number of impor-
tant cultivation successes, some of which have gar-
nered considerable far- reaching interest from different
fields. In most cases, the interest lies in the novelty of
the microorganisms that were cultured, or because the
cultured microorganisms provided insights and an
improved understanding of certain natural processes.
Among archaea, a notable example is the first
representative of the Asgard archaea superphylum,
‘Candidatus Prometheoarchaeum syntrophicum’,
representing the closest archaeal relative of eukary-
otes cultured to date, which was highly enriched in
a co- culture containing two species. This feat was
achieved using an innovative bioreactor system and
traditional enrichment methods over the course of
12 years, partially owing to this organism having
extremely slow growth rates11. The first representa-
tive of the Nanohaloarchaeota phylum, ‘Candidatus
Nanohaloarchaeum antarcticus’, was recently co- cultured
with a Euryarchaeota host, Halorubrum lacusprofundi,
by combining classical enrichment methods with
single- cell sorting selecting for cells of appropriate sizes,
as inferred from fluorescence insitu hybridization (FISH)
experiments12.Two archaea belonging to closely related
genera, ‘Candidatus Argoarchaeum ethanivorans’13
and‘Candidatus Ethanoperedens thermophilum’14,
are the first organisms shown to oxidize ethane in
Isolation
The physical separation of a
single cell, strain or species
from others found in the same
sample or habitat.
Fluorescence insitu
hybridization
(FISH). A method of labelling
cells with a fluorescent signal
by binding fluorophore-
coupled oligonucleotide
probes to complementary
target molecules (usually 16S
rRNA) in biological samples.
Probes can be designed to be
highly taxon- specific, making it
possible to taxonomically
identify microorganisms on the
single- cell level.
Target
microorganism
or groupa
Common
environments
Superphylum or
phylum
Reasons they are of interest for cultivation
Bacteria (cont.)
Most wanted taxa
from the Human
Microbiome
Project80
Human Assorted These are bacteria recognized based on 119 OTUs that have been prioritized
owing to their evolutionary distance from already characterized strains and
their frequency among healthy human- derived samples. Cultivation of such
microorganisms is thought to be essential for providing a better understanding
of human health and diseases80, and for example include bacteria belonging to
the phyla Bacteroidetes, Firmicutes and TM7.
SAR324 Marine Deltaproteobacteria They are metabolically diverse and globally distributed throughout the deeper
layers of the oceans144.
SAR86 Marine Gammaproteobacteria They are abundant in the surface layers of oceans and widespread globally145.
Most wanted taxa in
soil146
Soil Assorted These bacteria are thought to be crucial for accurately forecasting the ecological
consequences of ongoing global environmental change, and are important for
better understanding soil bacterial communities146. The most ubiquitous and globally
abundant of these include bacteria belonging to the phyla Alphaproteobacteria,
Betaproteobacteria, Actinobacteria, Acidobacteria and Planctomycetes.
Candidate Phyla
Radiation
Assorted Candidate Phyla
Radiation
This is a major group in Bacteria, currently thought to consist of at least 74
different phyla, for which there are an extremely limited number of cultured
representatives.
Most wanted taxa
in wastewater
treatment plants147
Wastewater
treatments
Assorted They are essential for municipal and industrial wastewater purification, by
removal of pollutants, to protect public and environmental health and have
importance for improving the performance of wastewater treatment plants147.
The most globally abundant and ubiquitous of these include bacteria belonging
to the phyla Betaproteobacteria, Gammaproteobacteria and Bacteroidetes.
Others
Most- wanted chem-
olithoautotrophic
‘spookmicrobes’148
Assorted Assorted These microorganisms from various taxonomic groups are thought to have
important roles in global methane, sulfur and nitrogen cycles. They also
participate in recently discovered processes, including complete ammonia
oxidation (comammox) and as yet undiscovered processes, such as iron and
manganese- dependent methane and ammonium oxidation148.
‘Candidatus
Parakaryon
myojinensis’149
Hydrothermal
vent
Unknown This microorganism is represented by an unusual microscopically investigated cell
without molecular data, which is interesting as it has structural similarities to both
prokaryotes and eukaryotes yet is seemingly distinct from both149.
aA subjective overview of microorganisms that could be considered as key targets for cultivation. Although this summary is far from exhaustive, given that every
researcher has different interests, the organisms presented here were selected owing to wide general interest in them or because they bear significant relevance to
particular scientific questions. The table is updated and expanded from previous work10.
Table 1 (cont.) | Key targets for cultivation
Nature reviews
|
Microbiology
Reviews
syntrophic interactions with sulfate- reducing bacteria,
and were cultured using traditional selective enrichment
methods.
Among bacteria, 79 different isolates belonging to
diverse lineages of Planctomycetes were recently cultured
using several traditional methods, including selec-
tive enrichment, antibiotic treatment and solid media
streaking combined with colony picking15. The first
freshwater representative of the widely abundant SAR11
Alphaproteobacteria clade, ‘Candidatus Fonsibacter
ubiquis’, was cultured by high- throughput dilution- to-
extinction in an oligotrophic medium16. Three species
belonging to the phylum Saccharibacteria (TM7) of
the Candidate Phyla Radiation (CPR), a broad clade
that has few cultured representatives, were isolated
together with their host Actinobacteria from human
saliva samples in the first demonstration of the reverse
genomics17 method. ‘Candidatus Manganitrophus nod-
uliformans’ is the first organism shown to be capable
of manganese oxidation in syntrophic interaction with
a betaproteobacterium, and was cultured using selec-
tive substrate enrichment and dilution- to- extinction18.
Casimicrobium huifangae, the first isolate of a novel
family in the Betaproteobacteria with the potential to
support improved understanding of and processing
in wastewater treatment plants, was cultured using
traditional methods19.
Although the examples mentioned above are far
from an exhaustive list of all microorganisms that were
successfully cultured in recent years, the total combined
strains that are currently maintained in culture in vari-
ous laboratories or culture collections around the world
represent only a miniscule fraction of the total microbial
diversity that exists.
The uncultured majority
The tree of life, arguably one of the most important con-
cepts in biology, has been vastly expanded with several
archaeal and bacterial groups of high taxonomic rank
over the past decades1,2,20. Contemporary best- supported
ideas for the structure of the tree of life divide prokary-
otes into two primary domains, Archaea and Bacteria,
which together are estimated to comprise anywhere
from hundreds to even thousands of phyla1,21–26 — a
figure that has increased as genome data have accu-
mulated, but can differ substantially depending on the
estimation method. Based on 16S rRNA gene sequence
data, the total number of archaeal and bacterial species
has been calculated to be around 400,000, comprising
around 60,000 genera22, although estimates of the actual
number of archaeal and bacterial species on Earth poten-
tially exceeds this by several orders of magnitude27–30.
However, only ~14,000 archaeal and bacterial species —
distributed over 3,500 genera and 38 phyla — have been
cultivated and validly described31–33. Of these species,
~97% belong to just four bacterial phyla (Bacteroidetes,
Proteobacteria, Firmicutes and Actinobacteria)31 (FIG.1).
Conversely, all other bacterial phyla, and Archaea as a
whole, are poorly represented by comparatively few
cultivated species (FIGS1,2).
Nevertheless, uncultivated or under- represented
phyla are known to dominate various environments,
where they are likely to have pivotal ecological roles34.
Therefore, cultivation of representative members of
these groups is important to uncover their physiolog-
ical and metabolic properties. Given the huge breadth
in diversity of microbial life, cultivating every microbial
species inhabiting our planet is practically impossible.
Therefore, in order to maximize effectiveness, attempts
to grow archaea and bacteria should prioritize represent-
atives of the most interesting or useful groups, or those
Dilution- to- extinction
A method of serially diluting a
mixed community culture with
the aim of isolating single cells
that will grow and divide to
establish monoclonal and
axenic cultures. Can also be
called limited dilution.
Box 1 | Classical cultivation strategies and methods
The origins of microbial cultivation can be traced back to the middle of the nineteenth
century, and many modern- day cultivation efforts rely on some of the same early principles
introduced more than a century ago154. Several strategies can be applied to enrich and
later isolate specific microorganisms, many of which rely on direct observation of the
physiological behaviour of the culture and the phenotypic and genotypic characteristics of
the microorganisms it contains (see the figure). Experience of the researcher with microbial
isolation is also important when it comes to the selection of the most appropriate measure
for isolation.
Examples of techniques to enrich specific taxa include the design of selective nutrient
media (for example, with specific substrates), application of selective physicochemical
conditions (for example, temperature, pH, salinity and gas- phase composition), addition
of selective inhibitors (for example, antibiotics, toxic compounds and metabolic inhibitors)
and the addition or omission of specific growth factors (for example, amino acids,
vitaminsand metals). The effect that each of these strategies has on the growth and
number of a specific population of microorganisms can be monitored and used to define
further isolation methods.
Observing cultures under the microscope can be a useful way to define strategies for
isolation. For example, when the target microorganism has a substantial difference in size
or shape from others in the culture, size fractionation by filters with various pore sizes
and mass- based separation by gradient centrifugation can be used to separate them.
Microscopic observation over time sometimes enables the detection of different growth
rates for microorganisms, which can then be used to inform subculture periods to select
for faster- growing microorganisms (by transferring cultures at an earlier incubation stage).
Growing cultures on a surface of solid media, commonly agar, and colony picking is a
common way to isolate organisms, and using alternative solidifying agents such as gellan
gum and agarose can target different microorganisms38,155. It is also possible to isolate
microorganisms in liquid media by dilution- to- extinction and design experiments for the
selection of motility phenotypes (such as phototaxis, aerotaxis, chemotaxis, galvanotaxis
or magnetotaxis)156–158.
Another consideration is the method used to sterilize growth medium, the most common
being autoclaving. However, besides the risk of degradation of certain components, the
presence of certain components during autoclaving can lead to the formation of toxic
by- products, such as hydrogen peroxide159, that can inhibit growth. Autoclaving media
components separately or, instead, using filter sterilization has been shown to avoid these
problems159.
Selective
nutrient media
Density-based
separation
Dilution-
to-extinction
Colony picking
from solid media
Selective
inhibitors
Size selection
Motility-based
enrichment
Varying
physicochemical
conditions
°C
pH
+
R
H
N
N
SCH3
CH3
O
O
H
OH
O
www.nature.com/nrmicro
Reviews
without any cultured representatives, and take advantage
of the most recent and innovative technologies to do so.
Factors influencing culturability
Identification of substrates and growth conditions. The
difficulties associated with the isolation and cultivation
of archaea and bacteria have long been recognized. In
the mid 1980s, 16S rRNA gene sequencing of environ-
mental samples revealed large numbers of uncultured
taxa35. From then on, the discrepancy between microor-
ganisms present in a given environment and those that
could be cultured in the laboratory was referred to as the
‘great plate count anomaly’36.
Since then, our understanding of microbial phys-
iology has improved substantially. The necessary sub-
strates, electron donors and acceptors, or other media
components for growing particular microorganisms can,
in some cases, be used to enrich and/or isolate specific
strains. However, cultivation of many microorganisms
strictly depends on specific environmental conditions
and on the presence of various growth factors (such as
vitamins, amino acids, nucleotides, inorganic com-
pounds, humic acids or other external electron shut-
tles), which are often difficult to identify and therefore
challenging to mimic in the laboratory37,38. Additionally,
several inorganic compounds (metals, sulfur and nitro-
gen compounds) are involved in cryptic cycles and
can be present at concentrations below the detection
limit in the environment, despite their importance in
biogeochemical cycles39.
Although for some microorganisms the broad cat-
egory of substrates that they use can be inferred from
genome sequences, it is possible that each uses a highly
specific subset of substrates from those categories, which
can be difficult to determine without functional data.
For example, genomic data suggest that Bathyarchaeota
and Asgard archaea commonly have heterotrophic
pathways for energy conservation and the ability to
degrade various organic substrates, including complex
carbohydrates, peptides, amino acids, alcohols, fatty
acids and hydrocarbons40,41. This variability could per-
haps partially explain why there is currently only one
cultured representative11 from these two diverse groups
of archaea.
Resuscitation of dormancy. Microbial populations can
include persisters, which are phenotypic variants of the
wild- type cells whose function is survival42. Persisting
microorganisms are dormant, non- dividing cells, and
in conditions of low nutrient and energy availability,
such as in the deep biosphere, dormancy might repre-
sent the default state of prokaryotic life43,44. As a large
proportion of microorganisms that currently remain
uncultured reside in environments that are potentially
dominated by persisters, resuscitation of dormancy
represents an essential hurdle in microbial cultivation
efforts. Although a substantial body of literature exists
on microbial dormancy, relatively little is known about
the potential mechanisms that underpin how micro-
organisms transition between dormant and active
states. Resuscitation of dormancy has been proposed
to be a stochastic process45, which might be influenced
by certain signalling compounds46,47. Still, microorgan-
isms have probably evolved different mechanisms to
regulate dormancy, which deems it unlikely that a uni-
fied solution exists to resuscitate them from dormant
states. Hence, this potential variability might represent
a further complication for cultivation studies.
Symbiotic interdependencies. In some cases, essential
molecules or electrons (including microbially pro-
duced electron shuttles, such as H2 and formate) are
directly exchanged between members of a microbial
community11,48–51, in an interspecies dependency com-
monly known as symbiosis (or ‘syntrophy’ if the two
organisms depend on each other for the degradation of
a substrate to overcome thermodynamic limitations).
In the case of mutualistic or syntrophic microorgan-
isms, using methods that can co- isolate both microbial
partners, such as cell sorting in a combinatorial fash-
ion, could be advantageous for establishing a stable
co- culture.
Given their (sometimes obligate) interdepend-
ence, separating symbiotic or syntrophic partners and
growing them in monocultures can be challenging.
However, attempts have been made to demonstrate
that one syntrophic partner can be abiotically replaced,
by investigating co- cultures of H2- producing bacte-
ria and H2- consuming methanogens52,53. In one study,
a H2- stripping bioreactor system was used to enrich
ethanol- oxidizing bacteria from a methanogenic
enrichment53. However, methanogenic activity was not
inhibited completely, suggesting that the H2 consump-
tion by the methanogen was not entirely replaced53.
Similar results were obtained in another study that used a
bioelectrochemical system to mimic H2 consumption by
methanogens in a co- culture, thereby greatly enriching
an ‘obligately’ syntrophic bacterium (Syntrophomonas
zehnderi)52. However, the bacterium was not separated
from the methanogens completely or maintained in a
monoculture.
Another option for the enrichment or isolation of
H2- producing syntrophs could be the catalytic removal
of H2. Previous studies have demonstrated the hydro-
genation of fatty acids using a palladium catalyst (both
fatty acids and H2 were produced through fermentation
of cellulose)54. However, hydrogenation rates in the liq-
uid phase of these experiments were low, and the effect
of the catalytically formed compounds on the growth of
H2- producing bacteria are not known, as this approach
was never tested for the purpose of microbial isolation.
Some microorganisms, such as the DPANN archaea and
the CPR bacteria, commonly have small genomes
and small cell sizes, which combined with an under-
standing of the lifestyles of the few cultured representa-
tives of these clades suggests that DPANN archaea and
CPR bacteria are predominantly dependent on other
host organisms to some degree, either in the form of
symbiosis or parasitism2. As such, this requirement
provides additional complications for cultivation as
appropriate conditions must be identified that satisfy
both microbial partners, which will likely require an
understanding of the basis for the relationship between
the partners.
Growth factors
Any substance that can be
used by an organism to
facilitate growth.
Symbiosis
The association, usually
a physical or metabolic
interaction, of two or more
organisms, which typically has
an influence on the fitness of
one or more of the partners
involved.
Syntrophy
An interspecies relationship in
which metabolites produced
by one species are used as
growth substrates by another
species.
Nature reviews
|
Microbiology
Reviews
Physical contact or spatial proximity. Physical contact
or spatial proximity between symbiotic or syntrophic
partners seems to be an important element for microbial
growth, as indicated by lower growth rates of microbial
partners grown in compartments separated by dialy-
sis membranes than for those grown together55, which
limits the application of such systems for microbial iso-
lation. Some researchers have applied Percoll gradient
techniques to separate syntrophic partners56,57, but trials
to further grow the segregated microorganisms often
fail, likely owing to the lack of growth factors typically
provided by other microorganisms. However, such seg-
regated fractions could still be used, for example, for
enzyme assays, giving further insights into the activity
of the separate microorganisms.
Physicochemical environmental conditions. Physico-
chemical environmental conditions, such as tem-
perature, pH, salinity and redox conditions, are also
important determinants for microbial cultivation, and
24 345 7
23
8
73
240
365
18
38
119
11*
11
6
8
4*
4
6
3
21
220
92
19
57
24
41
51
PVC superphylum
FCB group
Terrabacteria
Candidate Phyla Radiation
(Patescibacteria)
Scale: 0.1
24
Omnitrophica
(Omnitrophota)
51
Verrucomicrobia,
Lentisphaera and
Kiritimatiellaeota
(Verrucomicrobiota) 68
41
Planctomycetes
(Planctomycetota) 40
24
Elusimicrobia (Elusimicrobiota)
UBA6262
UBP18
Desantisbacteria
UBA9089
Firestonebacteria
Goldbacteria
FCPU426
3
Sumerlaeota
UBP4
OLB16
8
Nitrospinae (Nitrospinota)
UBA8248
Schekmanbacteria
7
Deferribacteres (Deferribacterota)
Chrysiogenetes (Chrysiogenetota)
Aquificae
(Aquificota)
Thermosulfidibacterota 1
11
4
27
45 Acidobacteria
(Acidobacteriota) 59
23
Nitrospirae (Nitrospirota)
CG2-30-53-67
Rokubacteria
(Methylomirabilota)
7
73
Oligoflexia
(Bdellovibrionota)
Deltaproteobacteria
(Myxococcota)
Deltaproteobacteria
(Desulfuromonadota)
Deltaproteobacteria
(Binatota)
Deltaproteobacteria
(Desulfobacterota)
8
40
214
61
MBNT15
GWC2
SZUA-79
O2-12
UBP6
UBA10199
240
Deltaproteobacteria
(Deferrisomatota)
UBA2233
Proteobacteria:
Alphaproteobacteria
Gammaproteobacteria
Zetaproteobacteria
Magnetococcia
2
1
1
1739
2990
18
Dependentiae
Lindowbacteria
WOR-3
38
Spirochaetes (Spirochaetota)
Deltaproteobacteria and
Epsilonproteobacteria
(Campylobacterota)
Chlamydiae (Verrucomicrobiota A)
UBP7
CG03
92
138
119
Bacteroidetes and Chlorobi
(Bacteroidota)
Marinimicrobia (Marinisomatota)
Calditrichaeota (Calditrichota)
Delongbacteria
KSB1
SM23
3
1572
11*
Latescibacteria (Latescibacterota)
Zixibacteria
TA06
4572-55
11
Fibrobacteres
(Fibrobacterota)
Gemmatimonadetes
(Gemmatimonadota)
3
4
6
Edwardsbacteria
TA06
Eisenbacteria
Krumholzibacteriota
8
Cloacimonetes
(Cloacimonadota)
Fermentibacteria
(Fermentibacterota)
WOR-3
4
Hydrogenedentes
(Hydrogenedentota)
UBA8481
GCA-001730085
4* Poribacteria
UBP7
3
Deinococcus-Thermus
(Deinococcota) 97
6
Fusobacteria (Fusobacteriota)
Wallbacteria
Muirbacteria
Riflebacteria
UBP17
42
21
Synergistetes (Synergistota)
Atribacteria (Caldatribacteriota)
Caldiserica (Caldisericota)
Coprothermobacterota
Dictyoglomi (Dictyoglomota)
Bipolaricaulota
Thermotogae (Thermotogota)
Thermodesulfobiota
25
1
1
2
3
48
57
Actinobacteria
(Actinobacteriota) 3559
19
Cyanobacteria*
Margulisbacteria
1
92
Annatimonadetes, WS-1
and Abditibacteriota
(Armatimonsdota)
Chloroflexi (Chloroflexota)
AD3 (Dormibacterota)
WPS-2 (Eremiobacteriota)
UBP13
UBP15
Firmicutes H
3
38
Tenericutes
(Firmicutes and Firmicutes A, B, C,
D, E, F and G)
Firmicutes (DTU030)
220
2738
365
www.nature.com/nrmicro
Reviews
all of these factors can vary sharply in natural envi-
ronments across microscale distances. In a microbial
community, some microorganisms contribute to mak-
ing the environmental conditions amenable for others,
which complicates the isolation of the microorganisms
that depend on these effects. Similarly, the cultivation
of strictly anaerobic microorganisms is also technically
demanding, in particular with modern high- throughput
techniques involving cell sorting and growth in micro-
titre plates (see below). Working in anaerobic tents or
glove boxes is often the most convenient solution for
enrichment and cultivation of anaerobic microorgan-
isms (for example, for plating and sorting of cells and
transferring enrichment cultures). However, aseptic
conditions can be difficult to maintain in these settings,
although the possibility of contamination can be reduced
by incorporating an air filtration unit.
Low abundance and competition. Many prokaryotes
exist in nature at low abundance in complex microbial
communities, yet may still exert substantial influences
on certain processes58. A possible reason for this is that
metabolic rates of substrate degradation and growth
rates are not necessarily linked. This means that, in some
cases, a low- abundance microorganism might metab-
olize a substrate at a faster rate than another, more abun-
dant, microorganism with higher growth rates found in
the same environment.
To have the best chance of isolating such rare, yet
ecologically relevant microorganisms, identifying
environments in which these cells are naturally pres-
ent at the highest relative abundance would benefit
further enrichment efforts. One way to identify such
environments is by analysing publicly available 16S
rRNA gene data, or generating such data denovo for
uncharacterized sites, to select the best locations to sam-
ple. However, even if a microorganism is obtained at a
high relative abundance, cultivation attempts can still
fail if faster- growing microorganisms are also present.
Faster- growing microorganisms have the potential to
quickly outcompete slow- growing target microorgan-
isms, meaning that even if a microorganism is initially
enriched in a sample, its relative abundance can soon
be diminished when both of these types of microorgan-
isms are co- inoculated. Such competition often happens
when cultivation media are supplemented with rich sub-
strates (for example, yeast extract or peptone) or when
easily fermentable substrates are used as carbon sources.
Likewise, some microorganisms can have a high affin-
ity for a particular substrate, consuming it efficiently
when the substrate is present in limited concentrations.
Thereby, these microorganisms can prevent the growth
of other microorganisms that are able to use the same
substrate but have a lower affinity for it59. To tackle these
problems, several recent methods17,60–63 focus on isolat-
ing single cells from environmental samples and using
these as inocula, rather than gradually enriching micro-
organisms from mixed communities. Additionally, for
oligotrophs that are poorly adapted to a nutrient- rich
environment, the use of low- nutrient media has proven
successful for their cultivation64,65.
An additional difficulty for the targeted cultiva-
tion of slow- growing microorganisms involves the
long timescales of research, which has both practical
and economic6 implications for researchers. Microbial
growth rates can be affected by suboptimal conditions
provided in the laboratory, and substantially differ from
the ‘natural’ growth rates in the environment. Although
growth rates might be improved by attempting to opti-
mize growth conditions, slow- growing microorgan-
isms might represent less appealing targets for many
researchers given the extended timescales and increased
associated research costs.
Innovative techniques
Most current methods that aim to increase the rate at
which microorganisms of interest are isolated broadly
follow at least one of two main strategies (FIG.3). They
either rely on scaling- up the number of cell isolations to
increase the chance of isolating a species that is interest-
ing (high- throughput isolation and cultivation), or aim
to selectively isolate organisms with specific functional
characteristics or that belong to a specific taxonomic
group (targeted isolation). Methods that fall into these
two categories (FIG.4) are described in the sections below.
Membrane diffusion- based cultivation. Our inability to
produce culture media that sufficiently replicate all of
the necessary growth factors present in natural habitats
remains a limitation for many cultivation experiments7,66.
With this limitation in mind, several cultivation tech-
nologies centre on the principle of physically separating
Anaerobic
An organism that grows in the
absence of molecular O2.
Inocula
Samples of microorganisms
introduced to fresh medium for
initiating the growth of a new
culture.
Fig. 1 | Cultured bacteria are biased towards Bacteroidetes, Proteobacteria,
Firmicutes and Actinobacteria. A phylogenetic species tree for bacteria, inferred
from concatenated alignments of a minimum of 5 out of a total 15 ribosomal proteins
per species, encoded by 1,541 bacterial genomes that were obtained from the Genome
Taxonomy Database21. Numbers in white font in coloured circles are the number of
individual taxa in each collapsed clade, and are also used to connect corresponding
taxa names to clades. Numbers in black font in white ellipses next to taxa names
indicate the total number of species- level cultured isolates described for those taxa,
based on the number of species type strains assigned to each clade that are present
in the BacDive database31 (last accessed 6 April 2020). Taxa without numbers have no
cultured isolates recorded in BacDive31. Numerous cultured representatives have been
reported in the scientific literature that are not represented in the numbers in this
figure, because cultures have not been officially described and/or deposited in culture
collections, and are therefore not included in BacDive31 (a comprehensive database
recording all cultured bacteria including those not officially described or deposited
in culture collections is currently lacking). The tree was generated from datasets
containing homologous proteins from the different species included, which were
aligned separately using MAFFT (L- INS- i)150 and the alignments for each protein then
concatenated, such that those proteins belonging to the same species were combined
to form a single sequence. Poorly conserved sites in the concatenated alignment were
removed using trimAl151 with the option - gt 0.5. A phylogeny was generated from this
trimmed alignment using the model LG + C60 + F + R10 in IQ- TREE152 with 1,000 ultrafast
bootstrap replicates153. Branches labelled with black dots have support values ≥95%.
Given the limited protein data set used to infer this phylogeny, in some cases the deeper
relationships between some species or groups may not reflect more widely accepted
relationships based on more in- depth and better supported analyses. Particularly,
Deinococcus- Thermus (Deinococcota) and Chlamydiae (Verrucomicrobiota A) do not
group with other lineages of Terrabacteria and the PVC superphylum, respectively.
*Although numerous cultured representatives for numerous cyanobacterial lineages
exist, they are particularly under- represented in BacDive31. Unlike most bacteria, and
owing to historical reasons, Cyanobacteria are mostly classified using the Botanical
code (that is, International Code of Nomenclature for algae, fungi and plants). As a
result, Cyanobacteria lack defined type strains and are therefore not extensively listed
in BacDive31, and a comprehensive database of existing Cyanobacteria cultures is lacking.
◀
Nature reviews
|
Microbiology
Reviews
cells, while allowing them limited contact with their
natural habitats. Typically, separation is achieved by
a filter or membrane, with a pore size small enough
to enable the diffusion of growth factors but not cells.
Inthese set- ups, cells maintain access to essential growth
factors from their natural environments or syntrophic
partners, while replicating in isolation, forming ideally
axenic cultures or colonies66. Furthermore, potential
growth- suppressing metabolites produced by the iso-
lated microorganisms can freely diffuse away rather than
accumulate locally61,67. Mimicking environmental insitu
growth conditions in this way avoids the meticulous
development of suitable artificial conditions, in particu-
lar avoiding excessive quantities of nutrients typically
provided by classical media, which can be detrimental
to the growth of some species65.
One such method is the hollow- fibre membrane
chamber device61, which consists of numerous hollow
fibres (porous tubes) connected to syringes that func-
tion as cell isolation chambers. The fibres are each
inoculated with a single cell, by serially diluting envi-
ronmental cell suspensions, and are then submerged in
an environmental water sample that provides the cells
with growth factors. The hollow fibres are inoculated
using the syringes, which can also be used to subsam-
ple the chambers while still incubating the remaining
sample insitu.
Another membrane diffusion- based technology
with the capacity to perform high- throughput cul-
tivation experiments is the i(isolation)Chip60 (and
derivatives68,69), which consists of a plate accommodating
an array of small holes that function as micro- chambers
to capture and isolate single cells from environmental
samples. These holes are sealed with a membrane, and
the whole device is then incubated in the environment
from which the cells were originally sampled, providing
insitu conditions for isolated cells to grow. One notable
success of this technology was the discovery of a novel
antibiotic from bacteria inhabiting soil70.
The soil substrate membrane system71,72 (and related
methods73,74) is another membrane diffusion method
that specifically targets archaea and bacteria inhabiting
soils. This system involves dispensing cells from an envi-
ronmental (soil) extract on the upper side of a mem-
brane. The membrane is then placed on top of a sample
of the soil, which the cells can then access and use as
a growth substrate. The system is then incubated, ena-
bling clonal colonies to form, which can be screened for
99
13 5
10
13
17
3
411
248
718 76 46 2
11
78
35
3
66
4
99
Nanoarchaeota
Woesearchaeota
Pacearchaeota
Parvarchaeota
13
Nanohaloarchaeota
5
Aenigmarchaeota
17
Micrarchaeota
3
UAP2
35
Bathyarchaeota
3
Marsarchaeota
Korarchaeota
10
Altiarchaeota
13
Diapherotrites
(Lainarchaeota)
DPANN archaea
Euryarchaeota
Asgard
archaea
TAC K
archaea
7
Hydrothermarchaeota
411
Halobacteria
(Halobacterota) 271
248
Thermoplasmatota 11
18
Methanococci 16
46
Thermococci 37
76
Methanobacteria 47
Hadesarchaea
(Hadarchaeota) 2
Verstraetearchaeota
(Methanomethylicia) 4
Lokiarchaeota
Heimdallarchaeota
Thorarchaeota
11
78
Geoarchaeota
(Geoarchaeales)
Thaumarchaeota
(Nitrososphaeria) 4
66
Thermoproteales
Desulfurococcales
Sulfolobales 59
Scale: 0.1
Fig. 2 | Archaeal diversity is dominated by uncultured groups.
A phylogenetic species tree for archaea, inferred from concatenated
alignments of a minimum of 5 out of a total 15 ribosomal proteins per
species, encoded by 1,166 archaeal genomes that were obtained from the
Genome Taxonomy Database21. Numbers in white font in coloured circles
are the number of individual taxa in each collapsed clade, and are also used
to connect corresponding taxa names to clades. Numbers in black font in
white ellipses next to taxa names indicate the total number of species- level
cultured isolates described for those taxa, based on the number of species
type strains assigned to each clade that are present in the BacDive
database31 (last accessed 6 April 2020). Taxa without numbers have no
cultured isolates recorded in BacDive31. Numerous cultured representatives
have been reported in the scientific literature that are not represented in
the numbers in this figure, because cultures have not been officially
described and/or deposited in culture collections, and are therefore not
included in BacDive31. A comprehensive database recording all cultured
archaea including those not officially described or deposited in culture
collections is currently lacking. The tree was generated from datasets
containing homologous proteins from the different species included, which
were aligned separately using MAFFT (L- INS- i)150 and the alignments for
each protein then concatenated, such that those proteins belonging to the
same species were combined to form a single sequence. Poorly conserved
sites in the concatenated alignment were removed using trimAl151 with the
option - gt 0.5. A phylogeny was generated from this trimmed alignment
using the model LG + C60 + F + R10 in IQ- TREE152 with 1,000 ultrafast
bootstrap replicates153. Branches labelled with black dots have support
values ≥95%. Given the limited protein dataset used to infer this phylogeny,
in some cases the deeper relationships between some species or groups
may not reflect more widely accepted relationships based on more in- depth
and better supported analyses.
www.nature.com/nrmicro
Reviews
species of interest or to inoculate media for continued
cultivation. Other, similar techniques have isolated bac-
teria by growing them on the surface of filters floating on
liquid media, which enabled colonies to form for species
that did not grow on more conventional solid media75,76.
For larger- volume cultivations, a previous study77
described the diffusion bioreactor, which provides cells
access to their natural growth factors. This device con-
sists of an inner chamber containing inoculated growth
medium and an outer chamber containing a substrate,
such as soil or sediment, which are connected by sev-
eral holes sealed by membranes. This set- up enables
cells to grow and proliferate in conditions resembling
their natural environment before isolating single strains
using classical methods, such as dilution- to- extinction
and spread- plate colony picking (BOX1).
Many of these diffusion cultivation devices have been
used to facilitate the growth of phylogenetically novel
species, beyond those that were recovered from the
same environment using traditional cultivation methods
alone60,61,71,76,77.
Microfluidic systems for cultivation. Microfluidic sys-
tems are widely used for various biological research
applications, including cultivation. Generally, the ben-
efits of these systems include increased scalability, and
therefore throughput, by miniaturizing overall experi-
mental set- ups78, as well as the ability to manipulate large
numbers of single cells from environmental samples in
parallel and in the presence of a range of substrates or
under different physicochemical conditions78. In some
cases, these benefits can also be extended to the cultiva-
tion of anaerobic microorganisms, as some systems can
maintain low levels or an absence of O2 (REF.62).
One example of these microfluidic systems is the
SlipChip79, which was originally designed for use in
chemistry but was later repurposed for high- throughput
cultivation of bacteria62. The repurposed version works
by incubating single cells separately in thousands of
microcompartments, which can contain various media
and substrates. The microcompartments are formed by
co- aligned wells present in the interfacing surfaces of
two adjoining plates (that together form the chip). Once
inoculated, the chip is incubated, giving the cells time
to multiply and form micro- sized cultures. The plates
are then ‘slipped’ apart, dividing each microcompart-
ment in two, thereby forming two identical replicate
microcultures for each compartment. The replicates in
the wells of one plate can then be individually screened
for growth and/or taxonomic identification (which is
typically destructive) and the corresponding wells of the
other plate preserve live cells for continued cultivation.
Additionally, multiple SlipChips can also be used to
screen a range of different growth conditions in paral-
lel. To do this, the contents of all cell- containing wells
from one plate are pooled. Pooling is done separately for
multiple chips, which have all been incubated in sepa-
rate conditions. These pools are then genetically assayed
(for example, by PCR) to screen for the presence of spe-
cies of interest, indicating growth of that species under
a particular condition. The data this generates can be
used to narrow down the number of potential culture
conditions, thereby efficiently tailoring a suitable subset
for growing a target microorganism78. A previous study78
demonstrated the effectiveness of this approach for cul-
turing the first representative of a Ruminococcae genus
corresponding to one of the ‘most wanted’ taxa in the
Human Microbiome Project80.
As discussed in the sections above, many micro-
organisms rely on products of syntrophic partners
forgrowth. In such cases, the use of microfluidic chips
that grow single microorganisms in isolation, in fully
sealed chambers, likely prohibit successful cultivation.
Toovercome this limitation, a previous study designed
the nanoporous microscale microbial incubator system63,
which incorporates both microfluidic and membrane
diffusion- based technologies. Nanoporous microscale
microbial incubator chips comprise an array of thou-
sands of micro- scale diffusion chambers organized on
a microfluidic slide63. Once sealed, the chambers physi-
cally isolate individual cells, but facilitate the transfer of
a
b c
d
ATCGATACACGG
Terrestrial Freshwater Marine Host-associated
>Sequence
ATCGATTACAC
GGTACGACTGA
AGTTCGGACTT
GCGAGATCACT
GAGGTCGAACA
CATACTACGG
Fig. 3 | Workflows for isolating novel microorganisms for cultivation using high-
throughput or targeted approaches. a | Sequencing- based screening of habitats can
be used to identify locations with high relative abundance of target organisms, followed
by collection of cell samples from these sites. b | High- throughput approaches can be
achieved by inoculating media with single cells to establish large numbers of monocultures,
incubating cultures and then screening for growth, followed by screening of viable
cultures for those containing species of interest. c | Targeted approaches rely on isolation
of cells belonging to specific taxonomic or functional groups. d | Cultured isolates can be
used for downstream characterization and experimentation to investigate their biology.
Nature reviews
|
Microbiology
Reviews
growth factors and signalling compounds between all
cells in the slide by passive diffusion through the per-
meable chamber walls. Although this system has strong
potential for the isolation and cultivation of interesting
syntrophic organisms, there are currently no published
examples.
Finally, systems that encapsulate single cells or small
populations of cells, in either liquid81–84 or gel85,86 drop-
lets, are also commonly used for cultivation87,88. The
cell- containing droplets are typically manipulated and
incubated in a microfluidic device. These encapsulation
methods decrease competition between species, because
cells are grown in isolation from those in other drop-
lets, and have also been demonstrated in some cases to
recover more phylogenetically diverse microorganisms
than were recovered using traditional cultivation meth-
ods for the same samples84. Furthermore, as the number
of simultaneous cultivation experiments can be vastly
increased by manipulating millions of individual drop-
lets in parallel, these techniques offer extremely high
rates of experimental throughput89.
Isolation of cells by sorting and selection for taxonomy
or function. Cell- sorting technologies are a mainstay
for many areas of biological research, and can be used
to isolate single cells from cell suspensions of mixed
communities. Cell sorting can be performed with many
different technologies, some of which perform at high
speed, such as droplet- based90 and microfluidic- based91
sorters, that are available commercially. Other technol-
ogies, such as microscopically guided optical tweezers92,
can precisely manoeuvre and isolate single cells but with
a lower rate of throughput. Such an approach was used
for the isolation and co- cultivation of the nanosized
hyperthermophilic archaeon Nanoarchaeum equitans
and its host Ignicoccus hospitalis93. However, unlike
droplet sorters, which can expose cells to considerable
pressure, optical tweezers typically exert less pressure, so
are less detrimental (although, in some cases, they can
cause photodamage94). Many cell sorters can sortcells
stochastically, which is useful for high- throughput cul-
tivation experiments, as large numbers of single cells
can be distributed into separate wells containing growth
media. However, the likelihood of isolating particular
target cells can be increased by selectively sorting cells
based on detectable distinguishing phenotypes.
Fluorescence- activated cell sorting (FACS) is a common
method for sorting cells based on fluorescence signals.
Whereas some organisms have intrinsic fluorescence
properties (autofluorescence), some low- toxicity or
non- toxic fluorescent dyes can also be used to stain dif-
ferent cellular targets, such as DNA and phospholipid
membranes, making cells more distinguishable from
background levels of fluorescence. In addition, if only a
subset of species autofluoresce when excited at a certain
wavelength of light or are better stained than other spe-
cies by a particular dye, FACS could be used to separate
cells based on these properties, and enrich a fraction of
the community. However, more sophisticated labelling
methods can be applied to increase the taxonomic or
functional specificity of isolated microorganisms.
FISH is a widely used fluorescent labelling method,
which can be used to identify and quantify cells belong-
ing to specific taxonomic groups in a given sample95.
FISH- labelled cell samples can also be sorted with
FACS to enrich cells belonging to selected taxonomic
groups for sequencing- based studies96. In the vast major-
ity of FISH protocols, cell viability is not maintained,
because cells are chemically fixed and their membranes
permeabilized to give molecular probes access to their
intracellular target, while also maintaining the struc-
ture of the cells. However, a recent study97 has demon-
strated a ‘live- FISH’ method, which avoids cell fixation
and permeabilization, and instead incorporates probes
into living cells by chemical transformation. Using this
technique, living Alphaproteobacteria from natural
Optical tweezers
A method for isolating single
cells from cellular suspensions
by microscopy and laser
capture. Many optical tweezer
set- ups are now automated
and operate in microfluidic
chips. Cells are passed through
these chips in a suspension,
and those with a detectable
phenotype are captured,
relocated from the main flow to
a sterile outlet and collected.
g
e
n
o
m
i
c
s
Inner ring
Has been used to cultivate members of under-sampled or poorly characterized
groups with no or few cultured representatives
Has been used to cultivate new members of groups for which there were already
cultured representatives
Has only been applied in proof of principle experiments to cultivate or sort
members of mock communities
M
e
m
b
r
a
n
e
d
i
ff
u
s
i
o
n
-
b
a
s
e
d
c
u
l
t
i
v
a
t
i
o
n
C
e
l
l
s
o
r
t
i
n
g
-
b
a
s
e
d
c
u
l
t
i
v
a
t
i
o
n
M
i
c
r
o
fl
u
i
d
i
c
s
-
b
a
s
e
d
c
u
l
t
i
v
a
t
i
o
n
D
r
o
p
l
e
t
N
M
M
I
i
C
h
i
p
S
l
i
p
C
h
i
p
R
A
C
S
R
e
v
e
r
s
e
S
S
M
S
H
F
M
C
D
i
ff
u
s
i
o
n
L
i
v
e
-
F
I
S
H
c
u
l
t
i
v
a
t
i
o
n
b
i
o
r
e
a
c
t
o
r
Fig. 4 | Innovative methods for the isolation and cultivation of novel microorganisms.
Membrane diffusion- based cultivation methods (green), such as the i(isolation)Chip60,
hollow- fibre membrane chambers (HFMC)61, diffusion bioreactors77 or the soil substrate
membrane system (SSMS)71,72, use permeable membranes that enable nutrients and
metabolites to diffuse into the cultivation medium and thereby mimic more natural
conditions during cultivation. Microfluidics- based cultivation methods (blue), such
as nanoporous microscale microbial incubators (NMMI)63 or the SlipChip78,79, are able
to manipulate cells in small volumes and large numbers of replicates, and can also be
combined with various droplet cultivation methods87. Cell sorting- based techniques
(yellow), such as Raman- activated cell sorting (RACS)98,100, fluorescence insitu hybridization
of live cells (live- FISH)97 or reverse genomics17, provide a way to target a functional or
taxonomic subset of cells for isolation.
www.nature.com/nrmicro
Reviews
seawater were first labelled with FISH probes, sorted by
FACS and, subsequently, cultivated. Despite optimizing
the protocol to maximize cellular viability during the
live- FISH procedure, the best survival rates of cells in
this study were relatively low (1.24–2.82%, depending
on the strain tested), and therefore, in its current form,
live- FISH would most likely be unsuitable for isolating
microorganisms found in low abundance.
A recent technique that bridges the gap between
non- taxa- specific cellular stains that can label cells while
maintaining their viability and highly taxa- specific but
usually destructive FISH is reverse genomics17 (FIG.5).
This technique takes advantage of the ease with which
near- complete genome sequences for uncultured micro-
organisms can be reconstructed from environmental
samples using genome- resolved metagenomics. From
these genomes, membrane proteins with extracellularly
exposed domains that are conserved only among target
microorganisms are then predicted in silico, and used
as epitopes for the production of antibodies, which are
then tagged fluorescently. When used in complex envi-
ronmental cell suspensions, the raised antibodies should
bind to the matching protein epitopes of the target
cells, thereby marking the cells with fluorescent labels.
Provided the selected epitopes have low sequence con-
servation with other microorganisms, labelling should
be taxa- specific, enabling detection of target cells and
separation from the remaining sample by FACS. Single
cells labelled using this method were shown to retain
viability after being sorted and were successfully used as
inocula to establish new cultures17.
Raman- activated cell sorting98 offers an alternative to
fluorescence- based labelling and a way of isolating viable
cells, while selecting for those that are most active under
certain conditions, thereby corresponding to particular
ecological functions. For Raman- activated cell sorting,
cells are incubated in the presence of deuterium (D2O)
under growth conditions that are likely to favour the
activity of target microorganisms99. Deuterium is propor-
tionally incorporated into the synthesized lipids of more
active cells, thereby conferring those cells with a chemical
label99. The deuterium labels can then be detected using
Raman microspectroscopy in a microfluidic device, with
the corresponding cells then captured and immediately
sorted with optical tweezers100. The isolated cells can then
be used as inocula for downstream cultivation.
Limitations. Although many of the techniques discussed
above have potential for increasing rates of species iso-
lation, their success might not be consistent across the
existing diversity of microbial life. Theoretically these
methods could be applied to target many taxa, but in
some cases their practical application could be much
more challenging. For example, microorganisms that
form biofilms could be more challenging to separate for
cell sorting101. Likewise, many microorganisms are sam-
pled from environments, such as sands, soils, sediments
and faecal material, that contain non- biological parti-
cles that can interfere with molecular labelling and flow-
based methods. In these cases, cells need to be separated
from the particles before isolation, and procedures to
achieve separation are often not trivial101,102.
Phenotypes
The observable or detectable
traits of an organism influenced
by its genes (genotype) and
factors of its environment.
Fluorescence- activated cell
sorting
(FACS). The dispersion of cells
into separate containers, such
as test tubes or wells, based on
either natural or artificially
induced fluorescent properties
(for example, by fluorescent
stains or labelling techniques).
Genome- resolved
metagenomics
The reconstruction of genome
sequences from metagenomic
data, typically obtained
through bioinformatics
approaches in which contigs
from a single microorganism
are grouped (‘binned’)
together.
a b c
d
hi g f
e
Clades with cultured
representatives
Clades with no cultured
representatives
ATCGATACACGG
Fig. 5 | Reverse genomics for targeted isolation and cultivation of novel microorganisms. Reverse genomics17 can be
used for targeted cultivation of novel lineages. a | First, the target microorganism belonging to novel or important clades is
identified. b | The genome of the target microorganisms can be reconstructed from metagenomic data. c | Based on these
data, proteins can be predicted and highly expressed membrane proteins with extracellular domains can be identified.
d | This is followed by the synthesis of a target- protein domain antigen and inoculation into a suitable animal for antibody
production. e | The raised antibodies are then purified and coupled to a fluorescent dye. f | Antibodies are added to
environmental cell samples. g | The antibodies label the target cells. h | Cells can then be sorted by fluorescence- activated
cell sorting based on the antibody- conferred signal. i | Cells are sorted onto liquid or solid growth media. If the targets are
symbionts that physically associate with each other and if one cell is labelled, both microorganisms could be co- sorted
together and used to inoculate a syntrophic co- culture.
Nature reviews
|
Microbiology
Reviews
Another limitation of many of these methods, par-
ticularly if working with anaerobic microorganisms, is
that isolation experiments must be performed in anoxic
conditions. Although anaerobic chambers have been
used for traditional microbial culturing methods for
many decades, newer techniques such as those involv-
ing cell sorters typically require much larger equipment,
which is difficult to fit and operate in a typical anaerobic
chamber. The use of larger and more accessible cham-
bers could be a solution; another option is to adapt the
cell isolation equipment in such a way that cells can be
manipulated under anoxic conditions. Indeed, several
companies have developed fluorescence- based cell
sorters that are either small enough to fit into typical
anaerobic chambers or, otherwise, perform cell sorting
in an enclosed flow cell, which can be loaded into an
anaerobic chamber before being removed and loaded
into the sorter.
Another difficulty for high- throughput cultivation
methods is the supply of gaseous substrates, such as H2,
CO2, CO and CH4. Currently, devices to supply gas to
microtitre plates are unavailable, with the main chal-
lenge being the complete isolation of the headspace of
the individual wells. Furthermore, cultivation at high
temperatures has its own intrinsic challenge owing to
liquid evaporation, which is particularly problematic
when cultivating cells in a small volumes as evaporation
can lead to cultures drying out. In addition, evapora-
tioncan also cause condensation build- up inside well
lids or seals, which can prevent the monitoring of growth
by automatic optical density measurement.
It must also be recognized that the isolation of cells is
typically just the first part of a two- part puzzle. To then
culture a microorganism that has been isolated, main-
taining its growth continuously, a suitable medium and
physicochemical conditions must be found. One way of
finding optimal conditions is by inferring phenotypic
features from metagenome- assembled genomes (com-
plemented by proteomic and transcriptomic data) from
uncultured target microorganisms, and selecting the
medium and conditions based on this information103.
Numerous published strategies and tools are availa-
ble for estimating physiological and ecological traits,
such as the optimum growth temperature104 and anti-
biotic susceptibility105, from genome sequences. Such
strategies could help provide clues for ways in which a
micro organism can be successfully grown or enriched
in cultures. However, a genome sequence alone often
provides insufficient data for accurately determining
all necessary culture conditions to grow a particular
microorganism successfully.
As an alternative, or in addition to genome- based
methods, sophisticated ‘next- generation’ physiology
approaches106 can be used to more accurately determine
metabolic and physiological properties for target micro-
organisms in enrichments or environmental samples.
These approaches include techniques such as bioorthog-
onal non- canonical amino acid tagging, stable isotope
probing and the detection of substrates incorporated at
the single- cell level by technologies such as NanoSIMS,
Raman microspectroscopy and BrdU staining106.
Combining the insights that these methods provide with
careful observation of microorganisms in their natural
environments, while being particularly attentive to their
physical and metabolic traits, can inspire ingenuity and
help researchers to find a successful way of isolating
andculturing a particular microorganism. Likewise, com-
plementing modern innovative methods with traditional
methods will also help researchers to achieve cultivation.
Screening methods
For experiments that generate several enrichments, cul-
tures or colonies, one must determine which of these
enrichments contain cells that are actively growing and
which contain organisms of interest. If done efficiently,
such screening will help researchers determine cultures
to prioritize for further study, because maintaining vast
numbers of cultures, including those in which cells
are not viable or contain microorganisms of limited
interest, is costly and will decrease the overall effective-
ness of research. Screening is particularly important
for high- throughput cultivation experiments, which
therefore require equally high- throughput screen-
ing methods. The following sections outline various
suchmethods, but ultimately researchers must deter-
mine which methods are most appropriate for their
particular experimental setting.
Direct visualization. The observable formation of a col-
ony on solid media indicates the presence of active cells
and the use of commercial colony- picking robots can
increase the rate at which colonies are taxonomically
screened and reinoculated107. However, many micro-
bial strains grow very slowly on solid media108 or stop
growing after their colonies reach a small size109. These
‘microcolonies’ can be invisible to the human eye with-
out magnification. For cases such as these, it is unclear
whether current commercially available colony- picking
robots are sufficiently sophisticated and precise.
Therefore, without improvements to this technology,
manually picking microcolonies, while observing them
under a microscope, might be a more efficient method.
Alternatively, one technological solution avoids these
difficulties by growing single cells in individual liquid
droplets arranged on the surface of solid media, which
are easier to manipulate in an automated fashion81.
For experiments with inoculated liquid medium
cultures, visible turbidity can sometimes be observed,
indicating microbial growth. However, some archaea
and bacteria cultivated in the laboratory may only reach
such low maximum cell densities that they defy visible
detection by eye. Attempts to visualize cells by light
microscopy could also fail if cells are very small and/or
transparent, and can be laborious for a high- throughput
set- up. Cells can be made more conspicuous by staining
them with fluorescent live stains and visualizing them
with a fluorescence microscope; however, it is still diffi-
cult to confidently rule out microbial growth in a sample
if no cells are visualized.
Optical detection of growth. Photospectrometer plate-
readers can be used to perform optical density meas-
urements for liquid samples in separate wells of a
multiwell plate, thereby determining increased cell
Anoxic
A state of complete absence of
molecular O2, for example, in
an environment or a culture.
Optical density
A common spectrophotometric
method for assessing the cell
density of a liquid suspension,
typically by measuring the
extent at which light at a
600 nm wavelength is
scattered by cells as it passes
through a sample.
www.nature.com/nrmicro
Reviews
density indicating growth in a scalable, high- throughput
manner. However, this method can be unsuitable for
species that have a low per- cell density or that only grow
to low population densities. Optical density measure-
ments also do not provide an accurate indication of the
number of cells present in a liquid sample. An alternative
and more sensitive method is flow cytometry, which can
be used to efficiently screen for growth and also quantify
cell numbers from as little as tens of microlitres of a cul-
ture, thereby retaining greater volumes for further exper-
iments. Many flow cytometers are automated so that they
can screen and quantify several cultures grown in sepa-
rate wells of multiwell plates, making them suitable for
processing large numbers of liquid samples such as those
generated by high- throughput dilution- to- extinction
experiments110.
PCR and sequencing- based screening. Once viable
colonies or cultures in a large- scale experiment are
identified, they can be screened for species of interest.
If a limited number of species are targeted, perform-
ing PCR with primers specific for target species is an
effective and scalable screening method. In some cases,
PCR can be performed using just a small subsample
of a culture as the input, with cells being lysed and
DNA released for amplification by the initial (typically
~95–98 °C) denaturation step of the PCR. However, more
sophisticated lysis methods are required for many robust
cell types.
For cultures containing naturally occurring PCR
inhibitors (many of which exist111), direct PCR often
is unsuccessful. Although the effects of some inhibi-
tors can be mitigated111, carefully designed controls are
needed to determine whether a negative result is really
caused by the absence of target microorganisms in a
sample (although this is inherently difficult to conclu-
sively determine) or whether PCR inhibition or failure
has occurred (a false negative). Alternatively, there are
vast numbers of protocols and commercial kits avail-
able to extract and purify DNA from cells, helping to
remove most inhibitors. However, a proportion of the
cell material or extracted DNA is typically lost during
most of these protocols, meaning they could be unsuit-
able for small and precious samples. Furthermore,
DNA extraction can be time consuming and costly at
a largescale.
Another widely used PCR- based identification
method is 16S rRNA gene amplicon sequencing, which
can also help to determine the relative abundance and
species diversity in a sample. 16S rRNA genes are present
in all archaeal and bacterial genomes, and contain several
variable regions that can be used for taxonomic discrimi-
nation, as well as highly conserved regions. Against these
conserved regions, ‘universal’ primers can be designed
that capture large swathes of the total known diver-
sity of archaea and bacteria, while also discriminating
between different species112. Indeed, various primer sets
targeting conserved regions of this gene are described in
the scientific literature and are widely used in diversity
studies113–115. The amplified PCR products from several
different samples can be sequenced in multiplex using
various contemporary high- throughput technologies.
The resulting data can be used to infer diversity and the
relative abundances from different samples. 16S rRNA
amplicon sequencing can therefore be particularly useful
to screen or continually monitor enrichment cultures in
which a mixed community is present116.
Amplicon sequencing can suffer from primer biases,
however, which in some cases lead to substantial por-
tions of the known microbial diversity being missed117.
Especially, microorganisms of as yet uncultivated phyla
such as the CPR and novel groups of archaea are fre-
quently overlooked by amplicon sequencing approaches
because insertions in their 16S rRNA genes25 or mis-
matches with commonly used primer sets impede
their detection117. If a specific taxonomic group is
being targeted for cultivation, primers that better cap-
ture the total breadth of diversity in that group could
be advantageousfor screening both sampling sites and
enrichments.
With the increased availability of cheap sequencing
technologies, 16S rRNA gene amplicon sequencing is
now quick and more affordable, which shortens the
duration between sampling and data analysis, mean-
ing that enrichment cultures can be monitored on finer
timescales. Although amplicon sequencing provides
insight into relative abundances, these measurements
do not provide absolute abundances or total cell quan-
tities. To achieve absolute measurements, sequencing
data can be complemented with cell enumeration data
generated (for example, by flow cytometry) for the same
sample, thereby providing a more complete understand-
ing of microbial community composition in a culture or
enrichment118,119.
Furthermore, one recently developed platform has
used affordable sequencing technologiesandmicro-
fluidics to automate DNA preparation and whole- genome
sequencing for screening large numbers of samples in
parallel, such as those generate by high- throughput cul-
tivation experiments120. This platform was also shown
to obtain high- quality genomic data for low biomass
samples, making it suitable for screening isolates that
grow to low population densities in cultures or form
microcolonies on solid media.
MALDI- TOF mass spectrometry. An alternative method
of taxonomic identification, for which media composi-
tion and the presence of inhibitors are less of a consider-
ation, is MALDI- TOF mass spectrometry. This method has
proven to be both fast and cost- effective for identifying
isolates while filtering out conspecifics and non- target
taxa. As MALDI- TOF mass spectrometry is highly
sensitive, only a relatively low cell mass is needed to
record a mass profile for taxonomic identification121.
Commercially available systems typically provide profile
databases that make identification possible at the genus
to species level for a query isolate. Currently, however,
these systems are mostly used to identify microorgan-
isms from clinical or food- associated environments, ren-
dering their databases unsuited for the identification of
taxa from other environments, as well as novel isolates122.
Therefore, to become a useful taxonomic identification
platform for large- scale cultivation experiments, data-
bases would need to be complemented with profiles of a
Flow cytometry
A technique used to detect and
count cells based on physical
or chemical properties.
MALDI- TOF mass
spectrometry
MALDI is an ionization
technique used in mass
spectrometric analysis based
on embedding samples in a
special matrix from which they
are desorbed by laser light.
The technique allows the
analysis of biomolecules and
organic molecules.
Nature reviews
|
Microbiology
Reviews
wide diversity of microorganisms. Broadening of data-
bases could be achieved by including profiles of novel
cultured taxa with confirmed identity (for example,
by 16S rRNA gene sequencing121). Although databases
are likely to improve in the future, another limitation
is that MALDI- TOF mass spectrometry identifica-
tion currently cannot be used for samples in which
several species are present, as these will not produce a
clearprofile.
Targets for culturing
What qualifies as an interesting microorganism differs
depending on the researcher, according to their own
interests and motivations. TABLE1 lists groups of micro-
organisms for which a strong case could be made that
it would be interesting to culture them, based on, for
example, their inferred functions and/or ecology, or
because they so far have no cultured representative.
Groups that are highly abundant in an environment
likely have an important role in the biogeochemistry of
that environment, as exemplified by Bathyarchaeota,
various Marine Group archaea, Acidobacteria, SAR202
and SAR86 (TABLE1). Therefore, such groups are inter-
esting to culture to help better understand their func-
tions. Identifying microorganisms that fit this profile
can be relatively straightforward, as abundances can be
estimated for particular groups from publicly available
16S rRNA gene amplicon data (for example, from data
in the Sequence Read Archive123) or can be generated for
individual environments denovo.
Microorganisms also could be deemed interesting to
culture if they belong to a large group with no, or few,
previously cultured representatives, with key examples
including the CPR bacteria and DPANN archaea (FIGS1,2;
TABLE1). Furthermore, microorganisms that shed light
on evolutionary processes owing to their proximity in
the tree of life to an important evolutionary event are
also interesting to culture. One clear example of this is
the Asgard archaea, in particular the Heimdallarchaeota,
which are thought to be the closest archaeal relatives
of eukaryotes124–126. Culturing further representa-
tives of this group will facilitate comparisons between
their cellular features and those of eukaryotes, thereby
potentially helping to establish the major evolutionary
changes that occurred during early eukaryotic evolution
(eukaryogenesis) (TABLE1).
Ultimately, in- depth knowledge and topic- specific
priorities will help researchers to identify the targets
that are likely to be the most rewarding for cultivation
efforts.
Conclusions
The wide- scale need for microbial isolation and cultiva-
tion has led to the development of numerous innovative
methods. Most of these methods adopt either a targeted
(for example, reverse genomics17, Raman- activated cell
sorting98 and live- FISH97) or a high- throughput (for
example, iChip60, SlipChip62 and nanoporous micros-
cale microbial incubators63) strategy to isolate cells
from communities and environments, although some
methods incorporate both strategies to varying degrees.
Some of these methods, such as reverse genomics17, have
proven to be viable and accessible options for bringing
interesting microorganisms belonging to poorly sampled
clades into culture, while circumventing the tradition-
ally long time spans that can be associated with cultur-
ing. Other approaches, however, although theoretically
appealing, have not yet been successful beyond isolat-
ing members of mock or low- complexity communities
(FIG.4). There are two main, generally opposing, possible
explanations for why this might be the case. For one,
these methods currently fail to overcome poorly under-
stood yet limiting biological processes (such as microbial
dormancy), and are therefore only capable of culturing
a certain range of microorganisms. Alternatively, the
other explanation is that these methods have not yet
been applied widely enough, or developed to the point
at which they will be most effective.
With current available methods, we are only able
to culture microorganisms that represent a small frac-
tion of the existing landscape of microbial diversity. To
culture new microorganisms, the further development
and maturation of advanced cultivation technologies
will be required. The innovative methods discussed
in this Review may well represent avenues for a future
revolution in successful cultivation efforts.
Published online xx xx xxxx
1. Hug, L. A. etal. A new view of the tree of life.
Nat. Microbiol. 1, 16048 (2016).
2. Castelle, C. J. & Banfield, J. F. Major new microbial
groups expand diversity and alter our understanding
of the tree of life. Cell 172, 1181–1197 (2018).
3. Mulder, A., Van de Graaf, A. A., Robertson, L. &
Kuenen, J. Anaerobic ammonium oxidation discovered
in a denitrifying fluidized bed reactor. FEMS Microbiol.
Ecol. 16, 177–183 (1995).
4. Nunoura, T. etal. A primordial and reversible TCA cycle
in a facultatively chemolithoautotrophic thermophile.
Science 359, 559–563 (2018).
5. Daims, H. etal. Complete nitrification by Nitrospira
bacteria. Nature 528, 504–509 (2015).
6. Thrash, J. C. Culturing the uncultured: risk versus
reward. mSystems 4, e00130–e00219 (2019).
7. Stewart, E. J. Growing unculturable bacteria.
J. Bacteriol. 194, 4151–4160 (2012).
8. Overmann, J., Abt, B. & Sikorski, J. Present and
future of culturing bacteria. Annu. Rev. Microbiol. 71,
711–730 (2017).
9. Tyson, G. W. & Banfield, J. F. Cultivating the
uncultivated: a community genomics perspective.
Trends Microbiol. 13, 411–415 (2005).
10. Carini, P. A. “Cultural” renaissance: genomics breathes
new life into an old craft. mSystems https://doi.org/
10.1128/mSystems.00092-19 (2019).
11. Imachi, H. etal. Isolation of an archaeon at the
prokaryote–eukaryote interface. Nature 577,
519–525 (2020).
This study is the first to culture a member of the
Asgard archaea group, the closest prokaryotic
relatives of the eukaryote nuclear lineage,
providing key insights into the evolutionary process
of eukaryogenesis.
12. Hamm, J. N. etal. Unexpected host dependency of
Antarctic Nanohaloarchaeota. Proc. Natl Acad. Sci. USA
116 , 14661–14670 (2019).
This study uses traditional enrichment techniques
and single- cell sorting to isolate a novel
Nanohaloarchaeota, belonging to the poorly
characterized DPANN archaea group.
13. Chen, S.-C. etal. Anaerobic oxidation of ethane by
archaea from a marine hydrocarbon seep. Nature
568, 108–111 (2019).
14. Hahn, C. J. etal. “Candidatus Ethanoperedens,” a
thermophilic genus of Archaea mediating the anaerobic
oxidation of ethane. mBio 11, e00600–e00620 (2020).
15. Wiegand, S. etal. Cultivation and functional
characterization of 79 planctomycetes uncovers their
unique biology. Nat. Microbiol. 5, 126–140 (2019).
16. Henson, M. W., Lanclos, V. C., Faircloth, B. C. &
Thrash, J. C. Cultivation and genomics of the first
freshwater SAR11 (LD12) isolate. ISME J. 12,
1846–1860 (2018).
17. Cross, K. L. etal. Targeted isolation and cultivation
of uncultivated bacteria by reverse genomics.
Nat. Biotechnol. 37, 1314–1321 (2019).
This study demonstrates a novel targeted isolation
method, using taxa- specific antibody labelling
combined with FACS sorting, to culture a bacterium
from a poorly characterized bacterial phylum with
few cultured representatives.
18. Yu, H. & Leadbetter, J. R. Bacterial
chemolithoautotrophy via manganese oxidation.
Nature 583, 453–458 (2020).
19. Song, Y. etal. Casimicrobium huifangae gen. nov.,
sp. nov., a ubiquitous “most- wanted” core bacterial
taxon from municipal wastewater treatment plants.
Appl. Environ. Microbiol. 86, 1038 (2019).
20. Parks, D. H. etal. Recovery of nearly 8,000
metagenome- assembled genomes substantially
www.nature.com/nrmicro
Reviews
expands the tree of life. Nat. Microbiol. 2, 1533–1542
(2017).
21. Parks, D. H. etal. A standardized bacterial taxonomy
based on genome phylogeny substantially revises the
tree of life. Nat. Biotechnol. 36, 996–1004 (2018).
22. Yarza, P. etal. Uniting the classification of cultured and
uncultured bacteria and archaea using 16S rRNA
gene sequences. Nat. Rev. Microbiol. 12, 635–645
(2014).
23. Castelle, C. J. etal. Genomic expansion of domain
Archaea highlights roles for organisms from new phyla
in anaerobic carbon cycling. Curr. Biol. 25, 690–701
(2015).
24. Rinke, C. etal. Insights into the phylogeny and coding
potential of microbial dark matter. Nature 499,
431–437 (2013).
25. Brown, C. T. etal. Unusual biology across a group
comprising more than 15% of domain Bacteria.
Nature 523, 208–211 (2015).
26. Baker, B. J., Lazar, C. S., Teske, A. P. & Dick, G. J.
Genomic resolution of linkages in carbon, nitrogen,
and sulfur cycling among widespread estuary
sediment bacteria. Microbiome 3, 1–12 (2015).
27. Curtis, T. P., Sloan, W. T. & Scannell, J. W. Estimating
prokaryotic diversity and its limits. Proc. Natl Acad.
Sci. USA 99, 10494–10499 (2002).
28. Lennon, J. T. & Locey, K. J. The underestimation of
global microbial diversity. mBio 7, e01298–e01316
(2016).
29. Locey, K. J. & Lennon, J. T. Scaling laws predict global
microbial diversity. Proc. Natl Acad. Sci. USA 113 ,
5970–5975 (2016).
30. Lennon, J. T. & Locey, K. J. More support for Earth’s
massive microbiome. Biol. Direct 15, e3000106
(2020).
31. Reimer, L. C. etal. BacDive in 2019: bacterial
phenotypic data for high- throughput biodiversity
analysis. Nucleic Acids Res. 47, D631–D636 (2019).
32. Garrity, G. M. NamesforLife BrowserTool takes
expertise out of the database and puts it right in the
browser. Microbiol. Today 37, 9. (2010).
33. Parte, A. C. LPSN — list of prokaryotic names with
standing in nomenclature (bacterio.net), 20 years on.
Int. J. Syst. Evol. Microbiol. 68, 1825–1829 (2018).
34. Lloyd, K. G., Steen, A. D., Ladau, J., Yin, J. & Crosby,
L. Phylogenetically novel uncultured microbial cells
dominate earth microbiomes. mSystems https://doi.
org/10.1128/mSystems.00055-18 (2018).
This study demonstrates that a substantial
fraction of the microorganisms on Earth
belong to phylogenetic groups without cultured
representatives.
35. Olsen, G. J., Lane, D. J., Giovannoni, S. J., Pace, N. R.
& Stahl, D. A. Microbial ecology and evolution:
a ribosomal RNA approach. Annu. Rev. Microbiol.
40, 337–365 (1986).
36. Staley, J. T. & Konopka, A. Measurement of insitu
activities of nonphotosynthetic microorganisms in
aquatic and terrestrial habitats. Annu. Rev. Microbiol.
39, 321–346 (1985).
37. Overmann, J. Principles of enrichment, isolation,
cultivation and preservation of prokaryotes.
Prokaryotes 1, 80–136 (2006).
38. Davis, K. E., Joseph, S. J. & Janssen, P. H. Effects of
growth medium, inoculum size, and incubation time on
culturability and isolation of soil bacteria. Appl. Environ.
Microbiol. 71, 826–834 (2005).
39. Kappler, A. & Bryce, C. Cryptic biogeochemical cycles:
unravelling hidden redox reactions. Environ. Microbiol.
19, 842–846 (2017).
40. Spang, A. etal. Proposal of the reverse flow model
for the origin of the eukaryotic cell based on
comparative analyses of Asgard archaeal metabolism.
Nat. Microbiol. 4, 1138–1148 (2019).
41. Zhou, Z., Pan, J., Wang, F., Gu, J.-D. & Li, M.
Bathyarchaeota: globally distributed metabolic
generalists in anoxic environments. FEMS Microbiol.
Re v. 42, 639–655 (2018).
42. Lewis, K. Persister cells, dormancy and infectious
disease. Nat. Rev. Microbiol. 5, 48–56 (2007).
43. Lennon, J. T. & Jones, S. E. Microbial seed banks: the
ecological and evolutionary implications of dormancy.
Nat. Rev. Microbiol. 9, 119–130 (2011).
44. Hoehler, T. M. & Jørgensen, B. B. Microbial life under
extreme energy limitation. Nat. Rev. Microbiol. 11,
83–94 (2013).
45. Epstein, S. The phenomenon of microbial
uncultivability. Curr. Opin. Microbiol. 16, 636–642
(2013).
46. Nichols, D. etal. Short peptide induces an
“uncultivable” microorganism to grow invitro.
Appl. Environ. Microbiol. 74, 4889–4897 (2008).
47. Dworkin, J. & Shah, I. M. Exit from dormancy in
microbial organisms. Nat. Rev. Microbiol. 8, 890–896
(2010).
48. Zengler, K. & Zaramela, L. S. The social network of
microorganisms—how auxotrophies shape complex
communities. Nat. Rev. Microbiol. 16, 383–390
(2018).
49. Stams, A. J. M. & Plugge, C. M. Electron transfer in
syntrophic communities of anaerobic bacteria and
archaea. Nat. Rev. Microbiol. 7, 568–577 (2009).
50. Hidalgo- Ahumada, C. A. P. etal. Novel energy
conservation strategies and behaviour of
Pelotomaculum schinkii driving syntrophic propionate
catabolism. Environ. Microbiol. 20, 4503–4511 (2018).
51. Lovley, D. R. Syntrophy goes electric: direct interspecies
electron transfer. Annu. Rev. Microbiol. 71 , 643–664
(2017).
52. Guzman, J. J. L., Sousa, D. Z. & Angenent, L. T.
Development of a bioelectrochemical system as
a tool to enrich H2-producing syntrophic bacteria.
Front. Microbiol. 10, 110 (2019).
53. Valentine, D. L., Reeburgh, W. S. & Blanton, D. C.
A culture apparatus for maintaining H2 at sub-
nanomolar concentrations. J. Microbiol. Methods 39,
243–251 (2000).
54. Mountfort, D. O. & Kaspar, H. F. Palladium- mediated
hydrogenation of unsaturated hydrocarbons with
hydrogen gas released during anaerobic cellulose
degradation. Appl. Environ. Microbiol. 52, 744–750
(1986).
55. Stieb, M. & Schink, B. Cultivation of syntrophic
anaerobic bacteria in membrane- separated culture
devices. FEMS Microbiol. Lett. 45, 71–76 (1987).
56. de Bok, F. A. M., Luijten, M. L. G. C. & Stams, A. J. M.
Biochemical evidence for formate transfer in
syntrophic propionate- oxidizing cocultures of
Syntrophobacter fumaroxidans and Methanospirillum
hungatei. Appl. Environ. Microbiol. 68, 4247–4252
(2002).
57. Beaty, P. S., Wofford, N. Q. & McInerney, M. J.
Separation of Syntrophomonas wolfei from
Methanospirillum hungatii in syntrophic cocultures by
using Percoll gradients. Appl. Environ. Microbiol. 53,
1183–1185 (1987).
58. Lynch, M. D. J. & Neufeld, J. D. Ecology and
exploration of the rare biosphere. Nat. Rev. Microbiol.
13, 217–229 (2015).
59. Janssen, P. H. Selective enrichment and purification of
cultures of Methanosaeta spp. J. Microbiol. Methods
52, 239–244 (2003).
60. Nichols, D. etal. Use of iChip for high- throughput
insitu cultivation of “uncultivable” microbial species.
Appl. Environ. Microbiol. 76, 2445–2450 (2010).
61. Aoi, Y. etal. Hollow- fiber membrane chamber
as a device for insitu environmental cultivation.
Appl. Environ. Microbiol. 75, 3826–3833 (2009).
62. Ma, L. etal. Individually addressable arrays of replica
microbial cultures enabled by splitting SlipChips.
Integr. Biol. 6, 796–805 (2014).
63. Ge, Z., Girguis, P. R. & Buie, C. R. Nanoporous
microscale microbial incubators. Lab Chip 16,
480–488 (2016).
64. Connon, S. A. & Giovannoni, S. J. High- throughput
methods for culturing microorganisms in very- low-
nutrient media yield diverse new marine isolates.
Appl. Environ. Microbiol. 68, 3878–3885 (2002).
65. Bartelme, R. P. etal. Influence of substrate
concentration on the culturability of heterotrophic
soil microbes isolated by high- throughput dilution-
to-extinction cultivation. mSphere 5, e00024-20
(2020).
66. Kaeberlein, T., Lewis, K. & Epstein, S. S. Isolating
“uncultivable” microorganisms in pure culture in a
simulated natural environment. Science 296,
1127–1129 (2002).
67. Dorofeev, A. G. etal. Approaches to cultivation of
“nonculturable” bacteria: cyclic cultures. Microbiology
83, 450–461 (2014).
68. Goordial, J. etal. Field sequencing and life detection in
remote (79°26′N) Canadian high Arctic permafrost
ice wedge microbial communities. Front. Microbiol. 8,
2594 (2017).
69. Pascual, J., García- López, M., González, I. &
Genilloud, O. Luteolibacter gellanilyticus sp. nov.,
a gellan- gum-degrading bacterium of the phylum
Verrucomicrobia isolated from miniaturized diffusion
chambers. Int. J. Syst. Evol. Microbiol. 67,
3951–3959 (2017).
70. Ling, L. L. etal. A new antibiotic kills pathogens without
detectable resistance. Nature 517, 455–459 (2015).
This study uses an iChip to culture novel
environmental microorganisms, leading to the
discovery of a new type of antibiotic, demonstrating
one of the functional benefits of culturing novel
microorganisms.
71. Ferrari, B. C., Binnerup, S. J. & Gillings, M. Microcolony
cultivation on a soil substrate membrane system
selects for previously uncultured soil bacteria. Appl.
Environ. Microbiol. 71, 8714–8720 (2005).
72. Svenning, M. M., Wartiainen, I., Hestnes, A. G. &
Binnerup, S. J. Isolation of methane oxidising bacteria
from soil by use of a soil substrate membrane system.
FEMS Microbiol. Ecol. 44, 347–354 (2003).
73. Zhao, H. etal. Different phenanthrene- degrading
bacteria cultured by insitu soil substrate membrane
system and traditional cultivation. Int. Biodeterior.
Biodegrad. 117 , 269–277 (2017).
74. Pudasaini, S. etal. Microbial diversity of Browning
Peninsula, Eastern Antarctica revealed using molecular
and cultivation methods. Front. Microbiol. 8, 591
(2017).
75. de Bruyn, J. C., Boogerd, F. C., Bos, P. & Kuenen, J. G.
Floating filters, a novel technique for isolation and
enumeration of fastidious, acidophilic, iron- oxidizing,
autotrophic bacteria. Appl. Environ. Microbiol. 56,
2891–2894 (1990).
76. Sipkema, D. etal. Multiple approaches to enhance the
cultivability of bacteria associated with the marine
sponge Haliclona (gellius) sp. Appl. Environ. Microbiol.
77, 2130–2140 (2011).
77. Chaudhary, D. K., Khulan, A. & Kim, J. Development
of a novel cultivation technique for uncultured soil
bacteria. Sci. Rep. 9, 6666 (2019).
78. Ma, L. etal. Gene- targeted microfluidic cultivation
validated by isolation of a gut bacterium listed in
Human Microbiome Project’s most wanted taxa.
Proc. Natl Acad. Sci. USA 111, 9768–9773 (2014).
This study uses the microfluidic SlipChip device to
culture the first member of the genus Ruminococcae,
which has relevance for better understanding the
human microbiome.
79. Du, W., Li, L., Nichols, K. P. & Ismagilov, R. F. SlipChip.
Lab Chip 9, 2286 (2009).
80. Fodor, A. A. etal. The “most wanted” taxa from the
human microbiome for whole genome sequencing.
PloS ONE 7, e41294 (2012).
81. Jiang, C.-Y. etal. High- throughput single- cell cultivation
on microfluidic streak plates. Appl. Environ. Microbiol.
82, 2210–2218 (2016).
82. Hu, B. etal. High- throughput single- cell cultivation
reveals the underexplored rare biosphere in deep- sea
sediments along the Southwest Indian Ridge. Lab Chip
20, 363–372 (2020).
83. Park, J., Kerner, A., Burns, M. A. & Lin, X. N.
Microdroplet- enabled highly parallel co- cultivation of
microbial communities. PloS ONE 6, e17019 (2011).
84. Watterson, W. J. etal. Droplet- based high- throughput
cultivation for accurate screening of antibiotic resistant
gut microbes. eLife 9, e56998 (2020).
This study demonstrates a high- throughput droplet
cultivation method that can isolate and grow large
numbers of single cells from environmental samples
anaerobically.
85. Zengler, K. etal. Cultivating the uncultured. Proc. Natl
Acad. Sci. USA 99, 15681–15686 (2002).
86. Eun, Y.-J., Utada, A. S., Copeland, M. F., Takeuchi, S.
& Weibel, D. B. Encapsulating bacteria in agarose
microparticles using microfluidics for high- throughput
cell analysis and isolation. ACS Chem. Biol. 6,
260–266 (2011).
87. Kaminski, T. S., Scheler, O. & Garstecki, P. Droplet
microfluidics for microbiology: techniques, applications
and challenges. Lab Chip 16, 2168–2187 (2016).
88. Boitard, L., Cottinet, D., Bremond, N., Baudry, J.
& Bibette, J. Growing microbes in millifluidic droplets.
Eng. Life Sci. 15, 318–326 (2015).
89. Najah, M. etal. Droplet- based microfluidics
platform for ultra- high-throughput bioprospecting
of cellulolytic microorganisms. Chem. Biol. 21,
1722–1732 (2014).
90. Leung, K. etal. A programmable droplet- based
microfluidic device applied to multiparameter analysis
of single microbes and microbial communities. Proc.
Natl Acad. Sci. USA 109, 7665–7670 (2012).
91. Shields, C. W. IV, Reyes, C. D. & López, G. P.
Microfluidic cell sorting: a review of the advances in
the separation of cells from debulking to rare cell
isolation. Lab Chip 15, 1230–1249 (2015).
92. Wang, X. etal. Enhanced cell sorting and manipulation
with combined optical tweezer and microfluidic chip
technologies. Lab Chip 11, 3656–3662 (2011).
93. Huber, H. etal. A new phylum of Archaea represented
by a nanosized hyperthermophilic symbiont. Nature
417, 63–67 (2002).
Nature reviews
|
Microbiology
Reviews
94. Zhang, H. & Liu, K.-K. Optical tweezers for single cells.
J. R. Soc. Interface 5, 671–690 (2008).
95. Wagner, M. & Haider, S. New trends in fluorescence
insitu hybridization for identification and functional
analyses of microbes. Curr. Opin. Biotechnol. 23,
96–102 (2012).
96. Haroon, M. F. etal. In- solution fluorescence insitu
hybridization and fluorescence- activated cell sorting
for single cell and population genome recovery
Methods Enzymol. 531, 3–19 (2013).
97. Batani, G., Bayer, K., Böge, J., Hentschel, U. &
Thomas, T. Fluorescence insitu hybridization (FISH)
and cell sorting of living bacteria. Sci. Rep. 9, 18618
(2019).
98. Lau, A. Y., Lee, L. P. & Chan, J. W. An integrated
optofluidic platform for Raman- activated cell sorting.
Lab Chip 8, 1116–1120 (2008).
99. Berry, D. etal. Tracking heavy water (D2O) incorporation
for identifying and sorting active microbial cells. Proc.
Natl Acad. Sci. USA 112, E194–E203 (2015).
100. Lee, K. S. etal. An automated Raman- based platform
for the sorting of live cells by functional properties.
Nat. Microbiol. 4, 1035–1048 (2019).
101. Müller, S. & Nebe- von-Caron, G. Functional single- cell
analyses: flow cytometry and cell sorting of microbial
populations and communities. FEMS Microbiol. Rev.
34, 554–587 (2010).
102. Morono, Y., Terada, T., Kallmeyer, J. & Inagaki, F.
An improved cell separation technique for marine
subsurface sediments: applications for high–throughput
analysis using flow cytometry and cell sorting. Environ.
Microbiol. 15, 2841–2849 (2013).
103. Gutleben, J. etal. The multi- omics promise in context:
from sequence to microbial isolate. Crit. Rev. Microbiol.
44, 212–229 (2018).
104. Sauer, D. B. & Wang, D.-N. Predicting the optimal
growth temperatures of prokaryotes using only genome
derived features. Bioinformatics 35, 3224–3231
(2019).
105. Su, M., Satola, S. W. & Read, T. D. Genome- based
prediction of bacterial antibiotic resistance. J. Clin.
Microbiol. 57, e01405–e01418 (2019).
106. Hatzenpichler, R., Krukenberg, V., Spietz, R. L.
& Jay, Z. J. Next- generation physiology approaches
to study microbiome function at single cell level.
Nat. Rev. Microbiol. 18, 241–256 (2020).
This review describes a range of methods that
can be used to gain a better understanding of the
physiological functions that microorganisms have in
their natural environments, which can also be used
to inform cultivation strategies.
107. Tillich, U. M. etal. High- throughput cultivation and
screening platform for unicellular phototrophs. BMC
Microbiol. 14, 239 (2014).
108. Bassi, C. A. & Benson, D. R. Growth characteristics of
the slow–growing actinobacterium Frankia sp. strain
CcI3 on solid media. Physiol. Plant. 130, 391–399
(2007).
109. Tahon, G. & Willems, A. Isolation and characterization
of aerobic anoxygenic phototrophs from exposed soils
from the Sør Rondane Mountains, East Antarctica.
Syst. Appl. Microbiol. 40, 357–369 (2017).
110 . Thrash, J. C., Weckhorst, J. L. & Pitre, D. M. in
Hydrocarbon and Lipid Microbiology Protocols Vol.
39 Springer Protocols Handbooks (eds McGenity T. J.,
Timmis K. N. & Nogales B.) 57–78 (Springer, 2017).
111. Schrader, C., Schielke, A., Ellerbroek, L. & Johne, R.
PCR inhibitors — occurrence, properties and removal.
J. Appl. Microbiol. 113 , 1014–1026 (2012).
112 . Kim, M. & Chun, J. in New Approaches to Prokaryotic
Systematics Vol. 41 (eds Goodfellow, M., Sutcliffe, I.
& Chu, J.) 61–74 (Elsevier, 2014).
113 . Bahram, M., Anslan, S., Hildebrand, F., Bork, P. &
Tedersoo, L. Newly designed 16S rRNA metabarcoding
primers amplify diverse and novel archaeal taxa from
the environment. Environ. Microbiol. Rep. 11,
487–494 (2019).
114 . Parada, A. E., Needham, D. M. & Fuhrman, J. A. Every
base matters: assessing small subunit rRNA primers
for marine microbiomes with mock communities, time
series and global field samples. Environ. Microbiol.
18, 1403–1414 (2016).
115 . Apprill, A., McNally, S., Parsons, R. & Weber, L. Minor
revision to V4 region SSU rRNA 806R gene primer
greatly increases detection of SAR11 bacterioplankton.
Aquat. Microb. Ecol. 75, 129–137 (2015).
116 . Rettedal, E. A., Gumpert, H. & Sommer, M. O. A.
Cultivation- based multiplex phenotyping of human
gut microbiota allows targeted recovery of previously
uncultured bacteria. Nat. Commun. 5, 4714 (2014).
117 . Eloe- Fadrosh, E. A., Ivanova, N. N., Woyke, T. &
Kyrpides, N. C. Metagenomics uncovers gaps in
amplicon- based detection of microbial diversity. Nat.
Microbiol. 1, 1–4 (2016).
118 . Fiedler, C. J. etal. Assessment of microbial community
dynamics in river bank filtrate using high- throughput
sequencing and flow cytometry. Front. Microbiol. 9,
2887 (2018).
119 . Props, R. etal. Absolute quantification of microbial
taxon abundances. ISME J. 11, 584–587 (2016).
120. Kim, S. etal. High- throughput automated microfluidic
sample preparation for accurate microbial genomics.
Nat. Commun. 8, 1–10 (2017).
121. Dumolin, C. etal. Introducing SPeDE: high- throughput
dereplication and accurate determination of microbial
diversity from matrix- assisted laser desorption–
ionization time of flight mass spectrometry data.
mSystems https://doi.org/10.1128/mSystems.00437-19
(2019).
122. Santos, I. C., Hildenbrand, Z. L. & Schug, K. A.
Applications of MALDI- TOF MS in environmental
microbiology. Analyst 141, 2827–2837 (2016).
123. Kodama, Y., Shumway, M. & Leinonen, R. The Sequence
Read Archive: explosive growth of sequencing data.
Nucleic Acids Res. 40, D54–D56 (2012).
124. Zaremba- Niedzwiedzka, K. etal. Asgard archaea
illuminate the origin of eukaryotic cellular complexity.
Nature 541, 353–358 (2017).
125. Spang, A. etal. Complex archaea that bridge the gap
between prokaryotes and eukaryotes. Nature 521,
173–179 (2015).
126. Williams, T. A., Cox, C. J., Foster, P. G., SzöllŐsi, G. J.
& Embley, T. M. Phylogenomics provides robust
support for a two- domains tree of life. Nat. Ecol. Evol.
4, 138–147 (2020).
127. Bhattarai, S., Cassarini, C. & Lens, P. Physiology and
distribution of archaeal methanotrophs that couple
anaerobic oxidation of methane with sulfate reduction.
Microbiol. Mol. Biol. Rev. 83, e00074–e00118 (2019).
128. Vanwonterghem, I. etal. Methylotrophic
methanogenesis discovered in the archaeal phylum
Verstraetearchaeota. Nat. Microbiol. 1, 1–9 (2016).
129. Dombrowski, N., Lee, J.-H., Williams, T. A., Offre, P.
& Spang, A. Genomic diversity, lifestyles and
evolutionary origins of DPANN archaea. FEMS
Microbiol. Lett. 366, fnz008 (2019).
130. Zhang, C. L., Xie, W., Martin- Cuadrado, A.-B.
& Rodriguez- Valera, F. Marine Group II archaea,
potentially important players in the global ocean
carbon cycle. Front. Microbiol. 6, 1108 (2015).
131. Haro- Moreno, J. M., Rodriguez- Valera, F., López-
García, P., Moreira, D. & Martin- Cuadrado, A.-B.
New insights into Marine Group III Euryarchaeota,
from dark to light. ISME J. 11, 1102–1117 (2017).
132. Reji, L., Tolar, B. B., Smith, J. M., Chavez, F. P. &
Francis, C. A. Differential co- occurrence relationships
shaping ecotype diversification within Thaumarchaeota
populations in the coastal ocean water column. ISME J.
13, 1144–1158 (2019).
133. Kielak, A. M., Barreto, C. C., Kowalchuk, G. A.,
van Veen, J. A. & Kuramae, E. E. The ecology of
Acidobacteria: moving beyond genes and genomes.
Front. Microbiol. 7, 744 (2016).
134. Becraft, E. D. etal. Rokubacteria: genomic giants
among the uncultured bacterial phyla. Front.
Microbiol. 8, 2264 (2017).
135. Ghai, R., Mizuno, C. M., Picazo, A., Camacho, A. &
Rodriguez- Valera, F. Metagenomics uncovers a new
group of low GC and ultra- small marine Actinobacteria.
Sci. Rep. 3, 2471 (2013).
136. Liu, Y.-F. etal. Anaerobic hydrocarbon degradation in
candidate phylum ‘Atribacteria’ (JS1) inferred from
genomics. ISME J. 13, 2377–2390 (2019).
137. Nobu, M. K. etal. Phylogeny and physiology of
candidate phylum ‘Atribacteria’ (OP9/JS1) inferred
from cultivation- independent genomics. ISME J. 10,
273–286 (2016).
138. Ji, M. etal. Atmospheric trace gases support primary
production in Antarctic desert surface soil. Nature
552, 400–403 (2017).
139. Hawley, A. K. etal. Diverse Marinimicrobia bacteria
may mediate coupled biogeochemical cycles along
eco- thermodynamic gradients. Nat. Commun. 8, 1–10
(2017).
140. Brewer, T. E., Handley, K. M., Carini, P., Gilbert, J. A.
& Fierer, N. Genome reduction in an abundant and
ubiquitous soil bacterium ‘Candidatus Udaeobacter
copiosus’. Nat. Microbiol. 2, 1–7 (2016).
141. Major, D. W. etal. Field demonstration of successful
bioaugmentation to achieve dechlorination of
tetrachloroethene to ethene. Environ. Sci. Technol. 36,
5106–5116 (2002).
142. Okazaki, Y., Salcher, M. M., Callieri, C. & Nakano, S.-I.
The broad habitat spectrum of the CL500-11 lineage
(phylum Chloroflexi), a dominant bacterioplankton in
oxygenated hypolimnia of deep freshwater lakes.
Front. Microbiol. 9, 2891 (2018).
143. Mehrshad, M., Rodriguez- Valera, F., Amoozegar, M. A.,
López- García, P. & Ghai, R. The enigmatic SAR202
cluster up close: shedding light on a globally distributed
dark ocean lineage involved in sulfur cycling. ISME J.
12, 655–668 (2018).
144. Sheik, C. S., Jain, S. & Dick, G. J. Metabolic flexibility
of enigmatic SAR324 revealed through metagenomics
and metatranscriptomics. Environ. Microbiol. 16,
304–317 (2014).
145. Dupont, C. L. etal. Genomic insights to SAR86, an
abundant and uncultivated marine bacterial lineage.
ISME J. 6, 1186–1199 (2012).
146. Delgado- Baquerizo, M. etal. A global atlas of the
dominant bacteria found in soil. Science 359,
320–325 (2018).
147. Wu, L. etal. Global diversity and biogeography of
bacterial communities in wastewater treatment plants.
Nat. Microbiol. 4, 1183–1195 (2019).
148. in ‘t Zandt, M. H., de Jong, A. E., Slomp, C. P.
& Jetten, M. S. The hunt for the most- wanted
chemolithoautotrophic spookmicrobes. FEMS
Microbiol. Ecol. 94, fiy064 (2018).
149. Yamaguchi, M. etal. Prokaryote or eukaryote?
A unique microorganism from the deep sea.
J. Electron. Microsc. 61, 423–431 (2012).
150. Katoh, K. & Standley, D. M. MAFFT multiple sequence
alignment software version 7: improvements in
performance and usability. Mol. Biol. Evol. 30,
772–780 (2013).
151. Capella- Gutiérrez, S., Silla- Martínez, J. M. &
Gabaldón, T. trimAl: a tool for automated alignment
trimming in large- scale phylogenetic analyses.
Bioinformatics 25, 1972–1973 (2009).
152. Nguyen, L.-T., Schmidt, H. A., von Haeseler, A. &
Minh, B. Q. IQ- TREE: a fast and effective stochastic
algorithm for estimating maximum- likelihood
phylogenies. Mol. Biol. Evol. 32, 268–274 (2015).
153. Hoang, D. T., Chernomor, O., von Haeseler, A.,
Minh, B. Q. & Vinh, L. S. UFBoot2: improving the
ultrafast bootstrap approximation. Mol. Biol. Evol. 35,
518–522 (2018).
154. Starr, M. P., Stolp, H., Trüper, H. G., Balows, A. &
Schlegel, H. G. The Prokaryotes: A Handbook on
Habitats, Isolation and Identification of Bacteria
(Springer Science & Business Media, 2013).
155. Tamaki, H. etal. Comparative analysis of bacterial
diversity in freshwater sediment of a shallow eutrophic
lake by molecular and improved cultivation- based
techniques. Appl. Environ. Microbiol. 71, 2162–2169
(2005).
156. Wolfe, R., Thauer, R. & Pfennig, N. A ‘capillary
racetrack’ method for isolation of magnetotactic
bacteria. FEMS Microbiol. Ecol. 3, 31–35 (1987).
157. Goossens, H., Wauters, G., De Boeck, M., Janssens, M.
& Butzler, J. Semisolid selective- motility enrichment
medium for isolation of salmonellae from fecal
specimens. J. Clin. Microbiol. 19, 940–941 (1984).
158. Yu, H. S. & Alam, M. An agarose- in-plug bridge method
to study chemotaxis in the Archaeon Halobacterium
salinarum. FEMS Microbiol. Lett. 156, 265–269
(1997).
159. Tanaka, T. etal. A hidden pitfall in the preparation of
agar media undermines microorganism cultivability.
Appl. Environ. Microbiol. 80, 7659–7666 (2014).
Acknowledgements
The authors are grateful to H. Smidt for inspiring discussions
and to F. Homa for performing the phylogenetic analyses for
the trees depicted in Figs 1 and 2. This work was supported
by grants from the European Research Council (ERC consoli-
dator grant 817834), the Dutch Research Council (NWO- VICI
grant VI.C.192.016) and the Wellcome Trust foundation
(Collaborative award 203276/K/16/Z) to T.J.G.E.
Author contributions
The authors contributed equally to all aspects of the article.
Competing interests
The authors declare no competing interests.
Peer review information
Nature Reviews Microbiology thanks Slava Epstein, Hiroyuki
Imachi, Yoichi Kamagata and the other, anonymous, reviewer(s)
for their contribution to the peer review of this work.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
© Springer Nature Limited 2020
www.nature.com/nrmicro
Reviews
