Magnetic genes: Studying the genetics of biomineralization in magnetotactic bacteria

Abstract

Many species of bacteria can manufacture materials on a finer scale than those that are synthetically made. These products are often produced within intracellular compartments that bear many hallmarks of eukaryotic organelles. One unique and elegant group of organisms is at the forefront of studies into the mechanisms of organelle formation and biomineralization. Magnetotactic bacteria (MTB) produce organelles called magnetosomes that contain nanocrystals of magnetic material, and understanding the molecular mechanisms behind magnetosome formation and biomineralization is a rich area of study. In this Review, we focus on the genetics behind the formation of magnetosomes and biomineralization. We cover the history of genetic discoveries in MTB and key insights that have been found in recent years and provide a perspective on the future of genetic studies in MTB.

Full text

REVIEW
Magnetic genes: Studying the genetics of
biomineralization in magnetotactic bacteria
Hayley C. McCauslandID
1
, Arash KomeiliID
1,2
*
1Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, California, United
States of America, 2Department of Plant and Microbial Biology, University of California, Berkeley, Berkeley,
California, United States of America
*komeili@berkeley.edu
Abstract
Many species of bacteria can manufacture materials on a finer scale than those that are syn-
thetically made. These products are often produced within intracellular compartments that
bear many hallmarks of eukaryotic organelles. One unique and elegant group of organisms
is at the forefront of studies into the mechanisms of organelle formation and biomineraliza-
tion. Magnetotactic bacteria (MTB) produce organelles called magnetosomes that contain
nanocrystals of magnetic material, and understanding the molecular mechanisms behind
magnetosome formation and biomineralization is a rich area of study. In this Review, we
focus on the genetics behind the formation of magnetosomes and biomineralization. We
cover the history of genetic discoveries in MTB and key insights that have been found in
recent years and provide a perspective on the future of genetic studies in MTB.
Author summary
Open any biology textbook and you are likely to learn that bacteria—unlike the cells of
plants, animals, and other eukaryotes—do not contain organelles to compartmentalize
and facilitate cellular functions. However, over the past several decades, many different
bacterial organelles have been discovered. In this Review, we highlight magnetotactic bac-
teria (MTB), which are a group of organisms capable of producing organelles called mag-
netosomes where nano-sized crystals of magnetic material are synthesized and housed. In
order to understand how and why MTB form magnetosomes, it is important to study the
genes involved. Here, we lay out the history of genetic studies in MTB and more recent
discoveries about which genes are involved at each step in the process of magnetosome
formation and discuss where the field is headed.
Introduction
Bacteria, according to the canonical definition, do not have subcellular compartments for
organization or specialized functions. Yet microbiologists are becoming increasingly aware
that many bacteria do have organelles, some of which are capable of manufacturing biomateri-
als with specialized functions [1]. MTB present a particularly elegant example of the biological
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 1 / 19
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: McCausland HC, Komeili A (2020)
Magnetic genes: Studying the genetics of
biomineralization in magnetotactic bacteria. PLoS
Genet 16(2): e1008499. https://doi.org/10.1371/
journal.pgen.1008499
Editor: Sarah M. Strycharz-Glaven, Naval Research
Laboratory, UNITED STATES
Published: February 13, 2020
Copyright: ©2020 McCausland, Komeili. This is an
open access article distributed under the terms of
the Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Funding: AK was supported by a grant from the
National Institutes of Health (R35GM127114). The
funder had no role in the decision to publish or the
preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.

behaviors that are mediated by intracellular compartments [2]. MTB are a group of bacteria
spanning multiple phyla that can be found in aquatic environments all over the world [3–5],
where they inhabit low oxygen environments, and are most often found at the oxic–anoxic
interface in the water column or sediments [6]. MTB are characterized by their ability to form
organelles called magnetosomes—lipid-bounded compartments in which biomineralization of
magnetic crystals of magnetite (Fe
3
O
4
) and/or greigite (Fe
3
S
4
) occurs [3]. Magnetosomes align
in one or multiple chains along the cell, creating a magnetic dipole that allows MTB to pas-
sively align along Earth’s magnetic field lines [7]. This is thought to help MTB perform more
efficient chemotaxis and aerotaxis in the water column as their swimming behavior is
restricted to one dimension instead of a three dimensional run and tumble search: a process
called magneto-aerotaxis [6]. The regulated process of biomineralization has made MTB an
attractive area of study in basic and applied biology, geochemistry, and physics alike. A greater
understanding of the molecular processes needed to form magnetosomes will enhance studies
in each of these fields, as knowledge of what is occurring on a molecular scale can lend greater
precision to technical applications and provide systems-level information for studying the
large-scale impacts of MTB.
The ecological impact of MTB is one rising area of research in the field. In recent years, our
understanding of the diversity of MTB has expanded greatly. In the process of biomineraliza-
tion of magnetite or greigite, MTB take up large amounts of dissolved iron from the surround-
ing environment and sequester it in magnetosomes as iron crystals. As such, the role that MTB
play in iron cycling in both freshwater bodies and the ocean is potentially quite large [5,8].
Conservative estimates from Amor and colleagues indicate that estuarine and oceanic MTB
may take up anywhere from approximately 1% to 50% of dissolved iron inputs (approximately
9×10
8
kg per year) in their environments [8]. Having a greater understanding of the iron reg-
ulation strategies encoded within the genomes of MTB, as well as how iron is taken up and dis-
tributed to magnetosomes, is important for a more accurate picture of the role of MTB in their
aquatic environments.
In this same vein, fossils of magnetosomes can help us understand the environmental con-
ditions that were present when MTB originated and the evolution of life on Earth. Magneto-
fossils—currently dating back to approximately 1.9 Ga—may reflect changes that occurred in
sediments and the water column and have the potential to serve as an indicator of redox and
oxygen levels in ancient environments [9]. Based on phylogenetic analyses, the origins of
biomineralization may have occurred much earlier, in the mid-Archaean (approximately 3
Ga), when the ability to biomineralize may have provided an advantage in coping with reactive
oxygen species, avoiding harmful UV radiation, and/or navigating ferrous-iron gradients
[10,11]. Understanding the genetic factors behind the formation of magnetosomes that are
common across modern MTB can provide insight into the conditions and processes that were
present when the first MTB originated [12]. Genetic analysis may also uncover unknown func-
tions of magnetosomes and can hint at the conditions that were needed to produce ancient
magnetosomes.
Additionally, the use of MTB in various biotechnological applications is promising. Magne-
tosomes are currently being developed for use as magnetic resonance imaging (MRI) contrast
agents, drug delivery systems, hyperthermic and photothermic treatments for cancer, biore-
mediation of heavy metals, and other nanotechnologies [13–15]. In order to efficiently pro-
duce the large numbers of magnetosomes required in these applications, it is critical to
understand how magnetosomes are produced.
Taking up iron from the environment for biomineralization, producing phospholipid
membranes of a specific size, and aligning magnetosomes in a chain is a complex, tightly con-
trolled process—one that is interesting in itself but also provides more general insights into the
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 2 / 19

precise formation of organelles. MTB encode the genes necessary for these processes in mag-
netosome gene clusters (MGCs) [16,17]. The MGCs in the most well-studied model organisms
—Magnetospirillum magneticum AMB-1, M.gryphiswaldense MSR-1, and Desulfovibrio mag-
neticus RS-1 (Fig 1)—are structured as magnetosome gene islands (MAIs). In both AMB-1
and RS-1, the MAI is defined by repeat regions on either side of the large chromosomal region
[18,19]. Across related species, there is a large amount of genetic homology in the MAI [20].
The functions of many of the genes within MGCs have been investigated, but much remains
unknown in each of the model organisms, and there is even more to be discovered about other
species and phyla of MTB.
In this Review, we lay out the history of landmark genetic discoveries in revealing impor-
tant insights into the process of magnetosome and biomineral formation by MTB. We also
dive into more nuanced views of genetics that have come out in recent years in step with
advances in genetic techniques. Finally, we will address the next directions that the field is tak-
ing to learn more about the genetics of MTB.
Genetics in Magnetospirilla
The discovery of MTB [7] fueled broad interest in understanding and exploiting the process of
organelle formation and biomineralization. Development of genetic systems greatly acceler-
ated the discovery of the molecular basis of magnetosome formation at a refined level. M.mag-
netotacticum MS-1 was the first magnetotactic bacterium to be isolated in pure culture [22].
However, MS-1 did not become a major model system for genetics since conditions that sup-
port colony growth on solid media have not been identified. Subsequently, AMB-1 and MSR-1
were successfully cultured and established as model organisms in the lab, making it possible to
manipulate and investigate their genomes [23–25].
Early studies of MTB genetics
The first genetic studies in MTB involved transposon mutagenesis coupled with magnetic
selection and transmission electron microscopy (TEM). Matsunaga and colleagues (1992) per-
formed mutagenesis in AMB-1 with the Tn5 transposon [23]. They identified several genomic
fragments involved in magnetosome synthesis by picking out mutant cultures that no longer
responded to a magnet under a light microscope. After confirming the mutants were deficient
in magnetosome formation with electron microscopy, they used restriction mapping to nar-
row down the location of each insertion site in the genome. One of these mutants carries a
transposon insertion in magA, a gene encoding a cation efflux pump proposed to function in
iron transport [26]. Importantly, this study also demonstrated that it was possible to transfer
plasmid DNA to AMB-1 using conjugation.
Large strides were made in understanding MTB genomes in the early 2000s. Wahyudi and
colleagues (2001) also isolated Tn5 transposon mutants in AMB-1 and found colonies with
defects in biomineralization by looking at colony color, which is thought to be an indicator of
how much magnetite has accumulated in the cell [27]. They concluded that at least 10, and up
to 60, genes could be involved in magnetosome formation. The publication of the genome
sequences of MS-1 and Magnetococcus marinus MC-1 also opened the door to genome-level
studies [28,29]. Gru¨nberg and colleagues (2001) compared protein sequences isolated from
MSR-1 magnetosomes to those in the MS-1 and MC-1 genomes and found two gene clusters
containing genes (mamA,mamB,mamC, and mamD) we now know to be critical for magne-
tosome formation [30].
The key genomic region needed for magnetosome formation, the MAI, was discovered
when spontaneous nonmagnetic mutants of MSR-1 were isolated from a wild-type population
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 3 / 19

of cells [17]. It was characterized as a 130-kb region containing multiple insertion sequence
(IS) elements [18]. The AMB-1 gene island was described as a 98-kb region flanked by two
1.1-kb repeat sequences [16]. A study of transcription of MAI genes indicated that while mag-
netosome genes are organized in operons, they are constitutively expressed [31]. Identifying
Fig 1. The MTB model systems. (A) TEM image of wild-type AMB-1 cell. (B) TEM image of wild-type RS-1 cell, scale bar 200 nm. Reprinted with permission from
Rahn-Lee and colleagues. [21]. TEM, transmission electron microscopy.
https://doi.org/10.1371/journal.pgen.1008499.g001
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 4 / 19

the MAI narrowed down the pool of genes to investigate and provided a foundation for more
targeted genetic studies.
In addition to defining the MAI, the establishment of MSR-1 and AMB-1 as model systems
allowed for more detailed molecular and genetic analyses [32,33]. A transposon mutagenesis
screen by Komeili and colleagues (2004) used a magnetic selection to enrich for nonmagnetic
mutants [34]. Colonies were then grown in 96-well plates and screened for magnetic response
using a 24-pin magnetic plate (Fig 2A). In this study, transposon insertions within the mamAB
gene cluster of the MAI resulted in nonmagnetic mutants. This work proved to be a great com-
plement to proteomic studies that had found the same MAI-encoded proteins associated with
magnetosomes [35].
Transposon mutagenesis studies proved to be a key turning point in using genetics to
understand the process of biomineralization in MTB. However, much like many other genetic
studies, their interpretation and broader utility were complicated by several confounding fac-
tors. First, due to homologous recombination between repeated sequences or potential action
of transposases, the MAI is unstable and can be lost spontaneously, an event that is more likely
to occur under stress conditions [18]. If transposition occurs in a bacterium that has lost its
MAI or if the MAI is lost following transposition, otherwise neutral events may appear linked
to changes in the magnetic phenotype. Screening potential mutants for the presence of the
MAI proved essential in isolating mutations in the mamAB region [34].
Second, many magnetosome genes contain functional paralogs that play redundant roles.
In AMB-1, three genes (mamQ,mamR, and, mamB) from the mamAB operon are perfectly
duplicated in another segment of the MAI [36]. Additionally, the AMB-1 genome contains a
magnetosome gene islet, a region outside of the MAI, which includes several homologs of
mamAB genes [37,38]. As a result, the absence of a distinct phenotype when any of the dupli-
cated genes are deleted individually does not rule out the possibility that they play a role in
magnetosome formation. Thus, in many cases, multiple genes must be deleted to understand
the function of a specific gene and its interactions with other genes. Additionally, complement-
ing deletions becomes critical for the evaluation of gene function.
Third, magnetosome genes are often organized as operons and transposon insertions result
in the polar loss of expression for all downstream genes. Thus, it is difficult to link a specific
phenotype to the loss of one single gene. Finally, by necessity, these studies used the magnetic
phenotype as a quick screening method to find relevant mutants. The secondary screens of
transposon mutants were important for establishing which step in the magnetosome forma-
tion process was affected by a particular mutation and allowed for assigning more specific
functions to genes (Fig 2B). For example, a nonmagnetic mutant might make magnetosome
membranes but not form crystals, indicating the interrupted gene was likely involved in crystal
formation. Or a nonmagnetic mutant might not make membranes at all, suggesting the site of
transposon insertion is a key part of magnetosome membrane formation.
Dissecting Magnetospirillum genomes
Obtaining the full genome sequences of the primary model organisms (MSR-1 and AMB-1)
propelled genetic investigations of MTB forward [33,39]. Previous studies provided limited
functional detail about any particular gene, at times involved looking at large deletions of mul-
tiple genes, and had complicating factors, as mentioned above. Deleting individual genes was
necessary for a more complete picture of magnetosome formation.
Murat and colleagues (2010) used previously developed methodology to thoroughly dissect
the MAI in AMB-1 by creating targeted deletions of genes and operons [36]. They began with
the observation that the loss of the MAI results in complete absence of both magnetosome
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 5 / 19

Fig 2. Magnetic screening technique. (A) (i) 24-pin magnetic plate (left) and 96-well plate of AMB-1 cells (right) used in Komeili and colleagues. [34]. (ii) Movement of
AMB-1 cells on magnetic plate at 0 seconds, 20 seconds, and 5 minutes. (iii) Phenotype of normal magnetic cells (left) and two representative, nonmagnetic mutants
(right). (B) Diagram of secondary screens to classify magnetosome mutants.
https://doi.org/10.1371/journal.pgen.1008499.g002
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 6 / 19

membranes and magnetic particles. Using a double recombination method for generating
nonpolar deletions, they first made mutants lacking larger subsections of the MAI [34,36].
Next, they focused on the regions that showed dramatic phenotypes such as small particles or
complete loss of the magnetosome membrane. Finally, they deleted individual genes within
these flagged regions and used a suite of secondary screens to assign specific functions to the
genes. Various electron microscopy techniques were used to visualize the magnetosome mem-
brane as well as the size, morphology, and subcellular arrangement of magnetic particles.
Green fluorescent protein (GFP) fusions to model magnetosome proteins were used to moni-
tor protein localization.
Through multiple layers of analysis, Murat and colleagues described the possible functions
of many of the key magnetosome-formation genes in AMB-1, like mamE,mamN,mamM,
mamO,mamI,mamL,mamQ, and mamB. Lohße and colleagues (2011, 2014) dissected the
MAI in MSR-1 [40,41] and found similar results, except that mamI and mamN were not essen-
tial for magnetosome formation in MSR-1. This discrepancy might be due to the particular
growth conditions used obscuring more subtle differences between the two species. It may also
reflect broader divergence between the two organisms. In a landmark study, heterologous
expression of the mamAB and mms6 operons, plus mamGFDC and mamXYZ, was found to be
sufficient to produce magnetosomes in the nonmagnetic α-Proteobacterium Rhodospirillum
rubrum [42], highlighting both the importance of these operons in magnetosome formation
and the minimal gene set needed to make magnetosomes under laboratory conditions.
These studies of the AMB-1 and MSR-1 islands provided a broad overview of the functions
of genes in the MAI. Further studies dove into investigating the functions of individual genes
in magnetosome formation and their mechanisms of action. The general framework that mag-
netosome genes are important for either membrane formation (Fig 3E) [43–45] or biomineral-
ization (Fig 3C and 3D) [46–53] holds true. A third category of genes that are involved in
chain organization has become an important area of study in recent years [38,54–59] (Fig 3A
and 3B). At the center of the chain-arrangement process is an actin-like protein called MamK.
Dynamic polymerization behavior of MamK is required for the integrity and proper segrega-
tion of the chain during cell division [54–58,60,61]. The proteins MamJ and MamY are also
key components of chain formation. MamJ acts as a link between MamK filaments and mag-
netosomes [54]. MamY is a cytoplasmic membrane protein that works to align the magneto-
some chain along the cell’s motility axis, which likely improves the efficiency of magnetotaxis
[59]. To add nuance to the broad categories, more specific functions of genes involved in each
stage of the process are being discovered as techniques and tools improve. More detailed sum-
maries of the genes and proteins responsible for magnetosome formation can be found in pre-
viously published Review articles (Fig 3F) [62–64].
In addition to the limits of techniques used in genetic analysis, the growth conditions of
MTB are an important factor to consider when examining the phenotype of particular genes.
Genes both inside and outside the MAI have been found to have important roles in biominer-
alization but only under certain conditions. For example, within the MAI ΔmamX,ΔmamZ,
ΔmamH, and ΔftsZm strains only show defects in biomineralization when MSR-1 is grown
with ammonium in place of nitrate, indicating that the use of oxygen instead of nitrate as the
terminal electron acceptor is detrimental to magnetosome formation [65–67]. Genes outside
the island, like the nap operon encoding nitrate-reductase genes [68] and the cytochrome c
oxidase cbb3 [69], are also key in the biomineralization process, again highlighting the impor-
tance of redox processes in magnetosome formation [70]. It is possible that these metabolic
pathways generate an overall redox balance within the cell that is compatible with biominerali-
zation, which requires both ferric and ferrous iron to be present. Alternatively, they may
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 7 / 19

participate directly in generating a redox-balanced iron pool that will allow for magnetite
biomineralization.
Another key environmental factor to consider in magnetosome formation is the availability
of iron. Unsurprisingly, genes involved in regulating the uptake of iron have been connected
to biomineralization. MSR-1 contains a homolog of the ferric uptake regulator (fur) gene com-
mon among bacteria [71]. This fur-like gene affects magnetosome size and number—poten-
tially due to reduced incorporation of iron into magnetite and an increase in cytoplasmic iron
concentrations—as well as transcription levels of several key MAI genes, including the
mamGFDC and mms6 operons [72]. Deletion of feoB1—which is involved in transport of fer-
rous iron into the cell—in MSR-1 resulted in fewer and smaller magnetosomes, as well as
decreased uptake of iron [73]. In AMB-1, the feoAB operon showed increased levels of tran-
scription under iron-rich conditions and was associated with an increase in intracellular iron
levels, suggesting that it is a component of iron uptake in AMB-1 [74]. Ferric iron transporters
were down-regulated in the same high-iron conditions. While the feoAB1 operon is within the
MAI, it is not a magnetosome-associated protein and thus its regulation of magnetosome
Fig 3. AMB-1 and MSR-1 strains with defects in magnetosome formation. (A) Wild-type, AMB-1–cell image taken from segmentation of an electron cryotomogram.
MamK filaments (green) run parallel to magnetosomes (yellow). (B) Electron cryotomogram image of a ΔmamK AMB-1 cell that shows disorganized magnetosomes.
Images provided by Komeili. (C) TEM image of a wild-type, AMB-1 cell. Scale bar is 0.2 μm. Close-up of magnetosomes is magnified 6×. (D) TEM image of a ΔmamT
AMB-1 cell showing small, misshapen magnetosomes. Scale bar is 0.2 μm. Close-up of magnetosomes is magnified 6×. Image provided by McCausland and colleagues.
(E) TEM image of a cryosection of ΔmamL AMB-1 cells showing that magnetosome membranes are absent. Scale bar is 0.2 μm. Image provided by Komeili. (F)
Diagram of the stepwise process of magnetosome formation and the proteins involved from membrane invagination (1), to crystal nucleation (2), and to membrane
growth and formation of a mature magnetic crystal (3). Genes that have been found to be involved at each step are listed. AMB-1, Magnetospirillum magneticum AMB-
1; IM, inner membrane; OM, outermembrane; MSR-1, Magnetospirillum gryphiswaldense MSR-1; TEM, transmission electron microscopy.
https://doi.org/10.1371/journal.pgen.1008499.g003
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 8 / 19

formation is indirect. MS-1 also expresses feoB, indicating that iron uptake systems are con-
served across species of MTB [75]. The iron response regulator IrrB was shown to be impor-
tant for magnetosome formation in MSR-1, and deletion affected the transcription of several
genes involved in the regulation of iron uptake [76]. These studies taken together indicate that
iron availability is an important factor to consider in magnetosome formation and genetic
regulation.
Genetics and genomics of diverse MTB
As described above, work in the model organisms MSR-1 and AMB-1 has identified many of
the genes that are important for the formation and positioning of magnetosomes. However,
MSR-1 and AMB-1 are both α-Proteobacteria species and MTB are an incredibly diverse
group of organisms with strains found in several classes of Proteobacteria, as well as Nitros-
pirae and the candidate division OP3 [3]. Little is known about the formation of magneto-
somes in nonmodel organisms. The genetic studies in MSR-1 and AMB-1 have been used as
jumping off points for the newer model system D.magneticus RS-1, as well as uncultured spe-
cies of MTB like those from other classes of Proteobacteria and the phyla Nitrospirae [20].
Additionally, improved sequencing technologies have allowed the field to study the phylogeny
of MTB using more relevant magnetosome genes instead of standard housekeeping genes.
The MAI in D.magneticus RS-1
The establishment of the δ-Proteobacterium RS-1 as a model system has opened up the field to
studying the genetic diversity of magnetosome formation. In the phylogeny of Proteobacteria,
the δ-Proteobacteria class is deeply branching, relative to α-Proteobacteria. As such, research
into δ-Proteobacteria can provide valuable insight into the origins of MTB. Through the study
of RS-1, it has been found that, in addition to a core set of genes required across all MTB, dif-
ferent types of MTB have distinct genes for magnetosome formation. Presumably the genes
evolved to adapt to the diverse lifestyles of each organism.
Comparative genome analysis of a variety of δ-Proteobacteria revealed that many of the
genes required for magnetosome formation in α-Proteobacteria are shared by the δ-Proteo-
bacteria, though some of the genes in the mamAB operon and the entire mamGFDC operon
are missing [19]. It was discovered in the same study that the δ-Proteobacteria and Nitrospirae
MTB have a separate set of class-specific genes termed the mad genes, which are likely involved
in the production of bullet-shaped magnetite crystals. Additionally, Nitrospirae MTB have
another set of genes, the man genes, that may be involved in the processes of magnetosome
formation and/or chain arrangement that are particular to Nitrospirae [77]. The magnetosome
gene sets that appear in different phyla provide a convincing link between genetic differences
and the clear phenotypic differences seen across MTB. Model organisms from the δ-Proteo-
bacteria and Nitrospirae are necessary to study the functions of mad and man genes.
In 1993, RS-1 was discovered as a sulfate-reducing MTB [78] and later identified as a Desul-
fovibrio species [79]. RS-1 is an obligate anaerobe that synthesizes bullet-shaped magnetite
crystals, as opposed to the cubo-octahedral crystals produced by the α-Proteobacteria MTB
species. While RS-1 was successfully cultured in the lab, attempts to delete individual genes
were unsuccessful at first. Rahn-Lee and colleagues got around this problem with a classic for-
ward genetic screen using random chemical and UV mutagenesis followed by whole-genome
sequencing of nonmagnetic mutants of RS-1 [21]. Both mam genes and mad genes were found
to be important in biomineralization, as were several novel MAI genes, including the ion
transport genes tauE and kup. Genetic tools for deleting specific genes in RS-1 were only
recently developed. Using suicide vectors for targeted gene deletion—as is commonly done in
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 9 / 19

other bacterial systems—does not work in RS-1 due to low transconjugation and recombina-
tion rates. Grant and colleagues developed a strategy using replicative plasmids that carry posi-
tive and negative selectable markers to replace the gene of interest with an antibiotic-resistance
gene [80].
The study of magnetosome formation in RS-1 also identified a novel organelle consisting of
iron–phosphorous granules surrounded by a membrane [81]. Byrne and colleagues showed
that these granules are separate organelles and not precursors to the formation of magnetite
with a pulse-chase experiment using different stable isotopes of iron. This conclusion was fur-
ther solidified by the finding that the deletion of the entire MAI of RS-1 had no impact on the
formation of the iron-rich granules [21]. The genetics of these novel bacterial organelles is a
rich area for future investigation.
The MAI in nonmodel and uncultured MTB
The genetic differences that have already been found between MSR-1, AMB-1, and RS-1 high-
light the need to study diverse MTB species in order to more fully understand magnetosome
formation and function(s). The rapid improvements in sequencing technologies in recent
years—in addition to the elegant method for isolating MTB using an external magnetic field—
have made it possible to study uncultured organisms in greater detail [82]. The genomes of
many uncultured MTB have been sequenced and analyzed in detail [19,28,42,77,83–85],
revealing that MTB belong to a wide variety of bacterial phyla. Metagenomic analyses are also
contributing greatly to our knowledge of the diversity of MTB and their evolutionary history
[10,20].
In the realm of nonmodel organisms, greigite-producing strains provide an interesting case
to study the evolution of MTB. Work by DeLong and colleagues analyzing 16S rRNA gene
sequences suggested that greigite-producing strains and magnetite-producing strains evolved
separately [86]. However, a later analysis by Abreu and colleagues found that the greigite-pro-
ducing strain Candidatus Magnetoglobus multicellularis has some of the mam genes that mag-
netite-producing strains require to produce magnetosomes, suggesting a monophyletic origin
for MTB [87]. Lefèvre and colleagues (2011) discovered the δ-Proteobacteria Desulfamplus
magnetovallimortis BW-1 and found that it is capable of producing both magnetite and greigite
[85]. Interestingly, the BW-1 genome has mam genes in two separate MGCs. Proteins encoded
in one cluster are closely related to proteins found in magnetite-producing species, while those
in the second cluster are more closely related to the proteins encoded in the MGCs of greigite
producers. The simplest hypothesis emerging from these genomic insights is that each cluster
is responsible for producing a chemically distinct, magnetic mineral. Lefèvre and colleagues
used the unique MGCs of BW-1 to examine phylogenetic differences between magnetite-pro-
ducing and greigite-producing strains [19]. Genes required for producing magnetite-contain-
ing magnetosomes are clustered together, as are those required for producing greigite-
containing magnetosomes, suggesting that there are separate sets of genes (and proteins)
involved in forming each type of crystal. However, the mad genes, which are needed to form
bullet-shaped magnetosomes, are present in both clusters. It is still unclear if mad genes were
lost during the evolution of magnetite-producing strains that do not form bullet-shaped crys-
tals or if they were acquired separately by δ-Proteobacteria and Nitrospirae strains of MTB.
Analysis of the α-Proteobacteria PR-1 also indicated that evolution of MTB likely involved
both vertical inheritance and horizontal gene transfer (HGT) or duplication events [84].
Looking further into the origins of MTB, Lefèvre and colleagues (2013) compared phyloge-
nies of several α-, δ-, and γ-Proteobacteria and one Nitrospirae MTB species. They constructed
phylogenetic trees using either 16S rRNA gene sequences and housekeeping genes or common
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 10 / 19

Mam proteins [88]. They found that both trees showed a similar pattern of divergence, leading
to the conclusion that all modern day Proteobacteria and Nitrospirae had a magnetotactic
common ancestor, though they did not rule out the possibility of an ancient HGT event. Two
recent studies from Lin and colleagues analyzed metagenomic data to gain insight into the ori-
gins of MTB [10,20]. The first study analyzed the genomes of multiple magnetotactic, Nitros-
pirae strains and found that the gene content and order in the MGCs were conserved across
the Nitrospirae, indicating a common origin. In the second study, a wide variety of MTB
genomes were analyzed using core magnetosome proteins and the phylogenetic trees showed
MTB clustering together. The authors concluded, like Lefèvre and colleagues, that HGT of
magnetosome genes were likely rare events. The simplest conclusion based on these studies is
that all MTB originated from a common ancestor. In fact, using commonly accepted molecular
clocks, it can be estimated that the original MTB—and presumably the first instance of magne-
tosome formation—appeared approximately 3.2 billion years ago [10]. An additional implica-
tion of this work is that at some point in the past the last common ancestor of the
Proteobacteria, Nitrospirae, and Omnitrophica phyla had the genes necessary for formation of
magnetic particles. The origins of magnetotactic Latescibacteria and Planctomycetes are less
clear. These phyla could have emerged from the last common ancestor of the magnetotactic
Proteobacteria, Nitrospirae, and Omnitrophica or acquired the genes through HGT. Subse-
quently, most descendants of these founding members lost the magnetosome genes leaving
behind the handful of modern-day MTB. The environmental conditions and changes that ini-
tially favored the evolution and expansion of magnetosome-formation genes and later selected
against them in the majority of bacteria remain to be elucidated. Perhaps, genetic studies of
other model MTB are needed to understand the potential contributions of group-specific
genes (such as mad and man genes) to the evolution and phenotypic diversification of
magnetosomes.
The study of uncultured MTB has also been aided through the analysis of gene function in
model MTB. Take, for example, the MAI genes mamE and mamO, which are critical for
biomineralization in both AMB-1 and MSR-1 [36,89]. Both gene products are predicted serine
proteases, and initial genetic studies concerning their functions concluded that this was indeed
the case [90]. However, further biochemical and structural studies revealed that the active site
of MamO is not functional and that it is in reality a metal-binding protein that controls
biomineralization and regulates the proteolytic activity of MamE [91,92]. Phylogenetic analy-
ses showed that, similar to AMB-1, all proteobacterial MTB encode an active and inactive pro-
tease in their MGCs [91]. The active protease is ancestral to all MTB and has been diversified
through vertical descent. However, the inactive protease has arisen multiple times in MTB
through duplications of the active protease or acquisition via HGT. These insights were only
possible through a combination of genetic, genomic, and biochemical studies. They highlight
the critical analyses needed in studies in which duplication events and diversification of func-
tion of similar proteins can blur the accuracy of phylogenetic studies. They also show that the
study of a protein in one species of MTB may not clarify the function of a homologous protein
in another related organism.
Outlook
Huge strides have been made in understanding the genetics behind magnetosome formation.
While the minimum set of genes required to generate magnetosomes is known, the specific
roles of many of these factors remain unknown. Additionally, there are several genes within
the MAI that when deleted have subtle or not obvious phenotypes. There are also presumably
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 11 / 19

many genes outside the island that have key, though indirect, roles in magnetosome
formation.
Genetic screens provide a high-throughput strategy for uncovering novel genes. Transpo-
son mutagenesis, in particular, has been used multiple times to study the genomes of AMB-1
and MSR-1. In the future, we envision several improvements that can make transposon muta-
genesis an even more useful method for genetic investigation of MTB. The latest techniques in
transposon mutagenesis involve pooling tens to hundreds of thousands of labeled mutants
with a method called random barcoded transposon-site sequencing (RB-TnSeq) [93] (Fig 4A).
This strategy allows for saturated coverage of a bacterial genome and averaged impact of gene
loss across multiple mutants. As such, polar effects and individual off-target effects are mini-
mized during phenotyping. However, genes identified in transposon-mutagenesis screens still
require phenotypic validation with gene deletions. The development of more advanced gene-
editing technologies, like CRISPR, will be just as valuable for the study of MTB as it has been
for other organisms [80,94]. For example, CRISPR interference (CRISPRi) can be used to
knockdown multiple genes at a time more readily than traditional methods or to study essen-
tial genes by tuning their expression [95,96]. Additionally, stepping aside from transposon
mutagenesis and screening instead for point mutants that have conditional phenotypes, act as
dominant alleles, or suppress known mutant phenotypes can help to expand our understand-
ing of the genetic networks that participate in magnetosome formation.
The methods for screening after the initial mutagenesis can also be refined. Screens thus far
have mostly relied on a simple binary identification of magnetic versus nonmagnetic cells or
have used colony color as a proxy for magnetite formation. The ability to identify mutants on a
spectrum of magnetic responses would be highly informative for understanding the process of
magnetosome formation (Fig 4B). It may be possible to use microfluidics for this approach
[97,98]. Methods that simply allow for the capture of more magnetic mutants are also key to
saturate screens and identify potential negative regulators of magnetosome formation.
Screening methods that allow for the identification of genes with more subtle phenotypes
on a spectrum will open the field to studying both genes in the MAI previously thought to
have little to no effect on magnetosome formation or genes outside the MAI that are key in
magnetosome formation under specific growth conditions. Multiple genes outside the MAI—
primarily involved in metabolism—have already been connected to magnetosome formation.
For example, the nitrate-reductase genes of the nap operon are important for magnetosome
formation in MSR-1, even when oxygen is available as the terminal electron acceptor [68].
And the metabolic regulator crp has also been tied to magnetosome formation [99]. Different
species of MTB migrate to a variety of preferred oxygen concentrations (all under 25 μM)
using one of three patterns of magneto-aerotaxis [100]. The genetic mechanisms behind aero-
tactic behavior have begun to be investigated. For example, Popp and colleagues showed that
the chemotaxis operon cheOp1 was necessary for the aerotactic response in MSR-1 [101].
The insights into metabolism and magnetosome formation are naturally connected to the
increasing interest in the field in studying the ecological role of MTB in their natural environ-
ments [102–104]. How MTB interact with and adapt to changing conditions will also be an
interesting problem from the perspective of geneticists and cell biologists. Most studies on
MTB have been done under tightly controlled laboratory conditions, but in nature, MTB
encounter changes in pH, temperature, oxygen gradients, and nutrient levels. A Review by
Moisescu and colleagues summarizes the effects of these changes on magnetosome formation
[105]. In addition, the study of environmental conditions may help us understand how
extremophile MTB evolved or retained the ability to form magnetosomes in conditions that
are not viable for most MTB species [106].
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 12 / 19

Fig 4. Barcoded transposon mutagenesis and a potential magnetic screen. (A) Diagram of RB-TnSeq in AMB-1. Each transposon insertion carries a unique
20-nucleotide sequence that acts as a barcode. The mutated strains are pooled, and then the barcodes are mapped to their insertion site in the genome. (B) Diagram of
magnetic selection with the TnSeq library using different magnetic strengths to select for a range of mutant phenotypes. From left to right, as magnetic strength applied
to the column increases, strains with weaker magnetic responses will be able to stick to the column, yielding a gradient of magnetic phenotypes to analyze. AMB-1,
Magnetospirillum magneticum AMB-1; BC1, barcode inserted into AMB-1 genome; gDNA, genomic DNA; N20, unique 20-nucleotide sequence; RB-TnSeq, random
barcoded transposon-site sequencing; U1, universal polymerase chain reaction priming site.
https://doi.org/10.1371/journal.pgen.1008499.g004
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 13 / 19

Conclusion
The set of genes needed for magnetosome formation has been clearly determined across multi-
ple organisms, and many of their functions have been investigated. However, it is clear that
even in the most well-studied MTB, like AMB-1 and MSR-1, the roles of genes that work in
tandem with other factors, that participate in multiple aspects of magnetosome formation, or
that are only required conditionally have yet to be fully understood. Going forward, more
nuanced study of genes involved in magnetosome formation will be key to expanding our
knowledge of MTB for basic cell biology, ecology, and biotechnology applications. Addition-
ally, the study of diverse MTB using both targeted genetic analyses and whole-genome studies
will potentially clarify the functions of many genes, while also adding layers to our picture of
MTB.
References
1. Grant CR, Wan J, Komeili A. Organelle Formation in Bacteria and Archaea. Annu Rev Cell Dev Biol.
2018 Oct 6; 34(1):217–38.
2. Rahn-Lee L, Komeili A. The magnetosome model: insights into the mechanisms of bacterial biominer-
alization. Front Microbiol. 2013; 4.
3. Lefèvre CT, Bazylinski DA. Ecology, Diversity, and Evolution of Magnetotactic Bacteria. Microbiol Mol
Biol Rev MMBR. 2013 Sep; 77(3):497–526. https://doi.org/10.1128/MMBR.00021-13 PMID:
24006473
4. Lin W, Bazylinski DA, Xiao T, Wu L-F, Pan Y. Life with compass: diversity and biogeography of magne-
totactic bacteria. Environ Microbiol. 2014; 16(9):2646–58. https://doi.org/10.1111/1462-2920.12313
PMID: 24148107
5. Lin W, Pan Y, Bazylinski DA. Diversity and ecology of and biomineralization by magnetotactic bacteria.
Environ Microbiol Rep. 2017; 9(4):345–56. https://doi.org/10.1111/1758-2229.12550 PMID:
28557300
6. Frankel RB, Bazylinski DA, Johnson MS, Taylor BL. Magneto-aerotaxis in marine coccoid bacteria.
Biophys J. 1997 Aug; 73(2):994–1000. https://doi.org/10.1016/S0006-3495(97)78132-3 PMID:
9251816
7. Blakemore R. Magnetotactic bacteria. Science. 1975 Oct 24; 190(4212):377–9. https://doi.org/10.
1126/science.170679 PMID: 170679
8. Amor M, Tharaud M, Ge
´labert A, Komeili A. Single-cell determination of iron content in magnetotactic
bacteria: implications for the iron biogeochemical cycle. Environ Microbiol. 2019. https://doi.org/10.
1111/1462-2920.14708 PMID: 31187921
9. Kopp RE, Kirschvink JL. The identification and biogeochemical interpretation of fossil magnetotactic
bacteria. Earth-Sci Rev. 2008 Jan 1; 86(1):42–61.
10. Lin W, Paterson GA, Zhu Q, Wang Y, Kopylova E, Li Y, et al. Origin of microbial biomineralization and
magnetotaxis during the Archean. Proc Natl Acad Sci U S A. 2017 28; 114(9):2171–6. https://doi.org/
10.1073/pnas.1614654114 PMID: 28193877
11. Lin W, Kirschvink JL, Paterson GA, Bazylinski DA, Pan Y. On the origin of microbial magnetorecep-
tion. Natl Sci Rev.
12. Jimenez-Lopez C, Romanek CS, Bazylinski DA. Magnetite as a prokaryotic biomarker: A review. J
Geophys Res Biogeosciences. 2010 Jun; 115(G2):n/a-n/a.
13. Yan L, Da H, Zhang S, Lo
´pez VM, Wang W. Bacterial magnetosome and its potential application.
Microbiol Res. 2017 Oct 1; 203:19–28. https://doi.org/10.1016/j.micres.2017.06.005 PMID: 28754204
14. Alphande
´ry E. Applications of magnetosomes synthesized by magnetotactic bacteria in medicine.
Front Bioeng Biotechnol. 2014; 2:5. https://doi.org/10.3389/fbioe.2014.00005 PMID: 25152880
15. Vargas G, Cypriano J, Correa T, Leão P, Bazylinski DA, Abreu F. Applications of Magnetotactic Bacte-
ria, Magnetosomes and Magnetosome Crystals in Biotechnology and Nanotechnology: Mini-Review.
Mol Basel Switz. 2018 Sep 24; 23(10).
16. Fukuda Y, Okamura Y, Takeyama H, Matsunaga T. Dynamic analysis of a genomic island in Magne-
tospirillum sp. strain AMB-1 reveals how magnetosome synthesis developed. FEBS Lett. 2006 Feb 6;
580(3):801–12. https://doi.org/10.1016/j.febslet.2006.01.003 PMID: 16423350
17. Schu¨bbe S, Kube M, Scheffel A, Wawer C, Heyen U, Meyerdierks A, et al. Characterization of a spon-
taneous nonmagnetic mutant of Magnetospirillum gryphiswaldense reveals a large deletion
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 14 / 19

comprising a putative magnetosome island. J Bacteriol. 2003 Oct; 185(19):5779–90. https://doi.org/
10.1128/JB.185.19.5779-5790.2003 PMID: 13129949
18. Ullrich S, Kube M, Schu¨bbe S, Reinhardt R, Schu¨ler D. A hypervariable 130-kilobase genomic region
of Magnetospirillum gryphiswaldense comprises a magnetosome island which undergoes frequent
rearrangements during stationary growth. J Bacteriol. 2005 Nov; 187(21):7176–84. https://doi.org/10.
1128/JB.187.21.7176-7184.2005 PMID: 16237001
19. Lefèvre CT, Trubitsyn D, Abreu F, Kolinko S, Jogler C, de Almeida LGP, et al. Comparative genomic
analysis of magnetotactic bacteria from the Deltaproteobacteria provides new insights into magnetite
and greigite magnetosome genes required for magnetotaxis. Environ Microbiol. 2013 Oct; 15
(10):2712–35. https://doi.org/10.1111/1462-2920.12128 PMID: 23607663
20. Lin W, Zhang W, Zhao X, Roberts AP, Paterson GA, Bazylinski DA, et al. Genomic expansion of mag-
netotactic bacteria reveals an early common origin of magnetotaxis with lineage-specific evolution.
ISME J. 2018 Jun; 12(6):1508. https://doi.org/10.1038/s41396-018-0098-9 PMID: 29581530
21. Rahn-Lee L, Byrne ME, Zhang M, Sage DL, Glenn DR, Milbourne T, et al. A Genetic Strategy for Prob-
ing the Functional Diversity of Magnetosome Formation. PLoS Genet. 2015 Jan 8; 11(1):e1004811.
https://doi.org/10.1371/journal.pgen.1004811 PMID: 25569806
22. Blakemore RP, Maratea D, Wolfe RS. Isolation and pure culture of a freshwater magnetic spirillum in
chemically defined medium. J Bacteriol. 1979 Nov; 140(2):720–9. PMID: 500569
23. Matsunaga T, Nakamura C, Burgess JG, Sode K. Gene transfer in magnetic bacteria: transposon
mutagenesis and cloning of genomic DNA fragments required for magnetosome synthesis. J Bacter-
iol. 1992 May; 174(9):2748–53. https://doi.org/10.1128/jb.174.9.2748-2753.1992 PMID: 1314800
24. Matsunaga T, Sakaguchi T, Tadakoro F. Magnetite formation by a magnetic bacterium capable of
growing aerobically. Appl Microbiol Biotechnol. 1991 Aug 1; 35(5):651–5.
25. Schleifer KH, Schu¨ler D, Spring S, Weizenegger M, Amann R, Ludwig W, et al. The Genus Magnetos-
pirillum gen. nov. Description of Magnetospirillum gryphiswaldense sp. nov. and Transfer of Aquaspir-
illum magnetotacticum to Magnetospirillum magnetotacticum comb. nov. Syst Appl Microbiol. 1991
Oct 1; 14(4):379–85.
26. Nakamura C, Burgess JG, Sode K, Matsunaga T. An Iron-regulated Gene, magA, Encoding an Iron
Transport Protein of Magnetospirillum sp. Strain AMB-1. J Biol Chem. 1995 Nov 24; 270(47):28392–6.
https://doi.org/10.1074/jbc.270.47.28392 PMID: 7499342
27. Wahyudi AT, Takeyama H, Matsunaga T. Isolation of Magnetospirillum magneticum AMB-1 mutants
defective in bacterial magnetic particle synthesis by transposon mutagenesis. Appl Biochem Biotech-
nol. 2001; 91–93:147–54. https://doi.org/10.1385/abab:91-93:1-9:147 PMID: 11963843
28. Schu¨bbe S, Williams TJ, Xie G, Kiss HE, Brettin TS, Martinez D, et al. Complete Genome Sequence
of the Chemolithoautotrophic Marine Magnetotactic Coccus Strain MC-1. Appl Env Microbiol. 2009 Jul
15; 75(14):4835–52.
29. Smalley MD, Marinov GK, Bertani LE, DeSalvo G. Genome Sequence of Magnetospirillum magneto-
tacticum Strain MS-1. Genome Announc. 2015 Apr 2; 3(2).
30. Gru¨nberg K, Wawer C, Tebo BM, Schu¨ler D. A large gene cluster encoding several magnetosome pro-
teins is conserved in different species of magnetotactic bacteria. Appl Environ Microbiol. 2001 Oct; 67
(10):4573–82. https://doi.org/10.1128/AEM.67.10.4573-4582.2001 PMID: 11571158
31. Schu¨bbe S, Wu¨rdemann C, Peplies J, Heyen U, Wawer C, Glo
¨ckner FO, et al. Transcriptional organi-
zation and regulation of magnetosome operons in Magnetospirillum gryphiswaldense. Appl Environ
Microbiol. 2006 Sep; 72(9):5757–65. https://doi.org/10.1128/AEM.00201-06 PMID: 16957191
32. Schultheiss D, Schu¨ler D. Development of a genetic system for Magnetospirillum gryphiswaldense.
Arch Microbiol. 2003 Feb; 179(2):89–94. https://doi.org/10.1007/s00203-002-0498-z PMID: 12560986
33. Matsunaga T, Okamura Y, Fukuda Y, Wahyudi AT, Murase Y, Takeyama H. Complete genome
sequence of the facultative anaerobic magnetotactic bacterium Magnetospirillum sp. strain AMB-1.
DNA Res Int J Rapid Publ Rep Genes Genomes. 2005; 12(3):157–66.
34. Komeili A, Vali H, Beveridge TJ, Newman DK. Magnetosome vesicles are present before magnetite
formation, and MamA is required for their activation. Proc Natl Acad Sci U S A. 2004 Mar 16; 101
(11):3839–44. https://doi.org/10.1073/pnas.0400391101 PMID: 15004275
35. Gru¨nberg K, Mu¨ller E-C, Otto A, Reszka R, Linder D, Kube M, et al. Biochemical and proteomic analy-
sis of the magnetosome membrane in Magnetospirillum gryphiswaldense. Appl Environ Microbiol.
2004 Feb; 70(2):1040–50. https://doi.org/10.1128/AEM.70.2.1040-1050.2004 PMID: 14766587
36. Murat D, Quinlan A, Vali H, Komeili A. Comprehensive genetic dissection of the magnetosome gene
island reveals the step-wise assembly of a prokaryotic organelle. Proc Natl Acad Sci U S A. 2010 Mar
23; 107(12):5593–8. https://doi.org/10.1073/pnas.0914439107 PMID: 20212111
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 15 / 19

37. Rioux J-B, Philippe N, Pereira S, Pignol D, Wu L-F, Ginet N. A second actin-like MamK protein in Mag-
netospirillum magneticum AMB-1 encoded outside the genomic magnetosome island. PLoS ONE.
2010 Feb 10; 5(2):e9151. https://doi.org/10.1371/journal.pone.0009151 PMID: 20161777
38. Abreu N, Mannoubi S, Ozyamak E, Pignol D, Ginet N, Komeili A. Interplay between two bacterial actin
homologs, MamK and MamK-Like, is required for the alignment of magnetosome organelles in Mag-
netospirillum magneticum AMB-1. J Bacteriol. 2014 Sep; 196(17):3111–21. https://doi.org/10.1128/
JB.01674-14 PMID: 24957623
39. Wang X, Wang Q, Zhang W, Wang Y, Li L, Wen T, et al. Complete Genome Sequence of Magnetospir-
illum gryphiswaldense MSR-1. Genome Announc. 2014 Mar 13; 2(2).
40. Lohße A, Borg S, Raschdorf O, Kolinko I, Tompa E
´, Po
´sfai M, et al. Genetic Dissection of the mamAB
and mms6 Operons Reveals a Gene Set Essential for Magnetosome Biogenesis in Magnetospirillum
gryphiswaldense. J Bacteriol. 2014 Jul; 196(14):2658–69. https://doi.org/10.1128/JB.01716-14 PMID:
24816605
41. Lohße A, Ullrich S, Katzmann E, Borg S, Wanner G, Richter M, et al. Functional Analysis of the Mag-
netosome Island in Magnetospirillum gryphiswaldense: The mamAB Operon Is Sufficient for Magne-
tite Biomineralization. PLoS ONE [Internet]. 2011 Oct 17; 6(10). Available from: https://www.ncbi.nlm.
nih.gov/pmc/articles/PMC3197154/. [cited 2019 Feb 7].
42. Kolinko I, Lohße A, Borg S, Raschdorf O, Jogler C, Tu Q, et al. Biosynthesis of magnetic nanostruc-
tures in a foreign organism by transfer of bacterial magnetosome gene clusters. Nat Nanotechnol.
2014 Mar; 9(3):193–7. https://doi.org/10.1038/nnano.2014.13 PMID: 24561353
43. Cornejo E, Subramanian P, Li Z, Jensen GJ, Komeili A. Dynamic Remodeling of the Magnetosome
Membrane Is Triggered by the Initiation of Biomineralization. mBio. 2016 Mar 2; 7(1):e01898–15.
https://doi.org/10.1128/mBio.01898-15 PMID: 26884433
44. Raschdorf O, Forstner Y, Kolinko I, Uebe R, Plitzko JM, Schu¨ler D. Genetic and Ultrastructural Analy-
sis Reveals the Key Players and Initial Steps of Bacterial Magnetosome Membrane Biogenesis. PLoS
Genet. 2016 Jun 10; 12(6):e1006101. https://doi.org/10.1371/journal.pgen.1006101 PMID: 27286560
45. Uebe R, Junge K, Henn V, Poxleitner G, Katzmann E, Plitzko JM, et al. The cation diffusion facilitator
proteins MamB and MamM of Magnetospirillum gryphiswaldense have distinct and complex functions,
and are involved in magnetite biomineralization and magnetosome membrane assembly. Mol Micro-
biol. 2011 Nov; 82(4):818–35. https://doi.org/10.1111/j.1365-2958.2011.07863.x PMID: 22007638
46. Murat D, Falahati V, Bertinetti L, Csencsits R, Ko
¨rnig A, Downing K, et al. The magnetosome mem-
brane protein, MmsF, is a major regulator of magnetite biomineralization in Magnetospirillum magneti-
cum AMB-1. Mol Microbiol. 2012 Aug; 85(4):684–99. https://doi.org/10.1111/j.1365-2958.2012.
08132.x PMID: 22716969
47. Siponen MI, Legrand P, Widdrat M, Jones SR, Zhang W-J, Chang MCY, et al. Structural insight into
magnetochrome-mediated magnetite biomineralization. Nature. 2013 Oct 31; 502(7473):681–4.
https://doi.org/10.1038/nature12573 PMID: 24097349
48. Rawlings AE, Bramble JP, Walker R, Bain J, Galloway JM, Staniland SS. Self-assembled MmsF pro-
teinosomes control magnetite nanoparticle formation in vitro. Proc Natl Acad Sci. 2014 Nov 11; 111
(45):16094–9. https://doi.org/10.1073/pnas.1409256111 PMID: 25349410
49. Kashyap S, Woehl TJ, Liu X, Mallapragada SK, Prozorov T. Nucleation of Iron Oxide Nanoparticles
Mediated by Mms6 Protein in Situ. ACS Nano. 2014 Sep 23; 8(9):9097–106. https://doi.org/10.1021/
nn502551y PMID: 25162493
50. Amemiya Y, Arakaki A, Staniland SS, Tanaka T, Matsunaga T. Controlled formation of magnetite crys-
tal by partial oxidation of ferrous hydroxide in the presence of recombinant magnetotactic bacterial pro-
tein Mms6. Biomaterials. 2007 Dec; 28(35):5381–9. https://doi.org/10.1016/j.biomaterials.2007.07.
051 PMID: 17720242
51. Yamagishi A, Tanaka M, Lenders JJM, Thiesbrummel J, Sommerdijk NAJM, Matsunaga T, et al. Con-
trol of magnetite nanocrystal morphology in magnetotactic bacteria by regulation of mms7 gene
expression. In: Scientific reports. 2016.
52. Arakaki A, Yamagishi A, Fukuyo A, Tanaka M, Matsunaga T. Co-ordinated functions of Mms proteins
define the surface structure of cubo-octahedral magnetite crystals in magnetotactic bacteria. Mol
Microbiol. 2014 Aug; 93(3):554–67. https://doi.org/10.1111/mmi.12683 PMID: 24961165
53. Jones SR, Wilson TD, Brown ME, Rahn-Lee L, Yu Y, Fredriksen LL, et al. Genetic and biochemical
investigations of the role of MamP in redox control of iron biomineralization in Magnetospirillum mag-
neticum. Proc Natl Acad Sci U S A. 2015 Mar 31; 112(13):3904–9. https://doi.org/10.1073/pnas.
1417614112 PMID: 25775527
54. Scheffel A, Gruska M, Faivre D, Linaroudis A, Plitzko JM, Schu¨ler D. An acidic protein aligns magneto-
somes along a filamentous structure in magnetotactic bacteria. Nature. 2006 Mar; 440(7080):110–4.
https://doi.org/10.1038/nature04382 PMID: 16299495
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 16 / 19

55. Komeili A, Li Z, Newman DK, Jensen GJ. Magnetosomes are cell membrane invaginations organized
by the actin-like protein MamK. Science. 2006 Jan 13; 311(5758):242–5. https://doi.org/10.1126/
science.1123231 PMID: 16373532
56. Draper O, Byrne ME, Li Z, Keyhani S, Barrozo JC, Jensen G, et al. MamK, a bacterial actin, forms
dynamic filaments in vivo that are regulated by the acidic proteins MamJ and LimJ. Mol Microbiol.
2011 Oct; 82(2):342–54. https://doi.org/10.1111/j.1365-2958.2011.07815.x PMID: 21883528
57. Toro-Nahuelpan M, Mu¨ller FD, Klumpp S, Plitzko JM, Bramkamp M, Schu¨ler D. Segregation of pro-
karyotic magnetosomes organelles is driven by treadmilling of a dynamic actin-like MamK filament.
BMC Biol. 2016 12; 14(1):88. https://doi.org/10.1186/s12915-016-0290-1 PMID: 27733152
58. Ozyamak E, Kollman J, Agard DA, Komeili A. The bacterial actin MamK: in vitro assembly behavior
and filament architecture. J Biol Chem. 2013 Feb 8; 288(6):4265–77. https://doi.org/10.1074/jbc.
M112.417030 PMID: 23204522
59. Toro-Nahuelpan M, Giacomelli G, Raschdorf O, Borg S, Plitzko JM, Bramkamp M, et al. MamY is a
membrane-bound protein that aligns magnetosomes and the motility axis of helical magnetotactic bac-
teria. Nat Microbiol. 2019 Jul 29;
60. Katzmann E, Scheffel A, Gruska M, Plitzko JM, Schu¨ler D. Loss of the actin-like protein MamK has
pleiotropic effects on magnetosome formation and chain assembly in Magnetospirillum gryphiswal-
dense. Mol Microbiol. 2010; 77(1):208–24. https://doi.org/10.1111/j.1365-2958.2010.07202.x PMID:
20487281
61. Taoka A, Kiyokawa A, Uesugi C, Kikuchi Y, Oestreicher Z, Morii K, et al. Tethered Magnets Are the
Key to Magnetotaxis: Direct Observations of Magnetospirillum magneticum AMB-1 Show that MamK
Distributes Magnetosome Organelles Equally to Daughter Cells. mBio. 2017 08; 8(4).
62. Uebe R, Schu¨ler D. Magnetosome biogenesis in magnetotactic bacteria. Nat Rev Microbiol. 2016 13;
14(10):621–37. https://doi.org/10.1038/nrmicro.2016.99 PMID: 27620945
63. Barber-Zucker S, Zarivach R. A Look into the Biochemistry of Magnetosome Biosynthesis in Magneto-
tactic Bacteria. ACS Chem Biol. 2017 20; 12(1):13–22. https://doi.org/10.1021/acschembio.6b01000
PMID: 27930882
64. Schu¨ler D. Genetics and cell biology of magnetosome formation in magnetotactic bacteria. FEMS
Microbiol Rev. 2008; 32(4):654–72. https://doi.org/10.1111/j.1574-6976.2008.00116.x PMID:
18537832
65. Raschdorf O, Mu¨ller FD, Po
´sfai M, Plitzko JM, Schu¨ler D. The magnetosome proteins MamX, MamZ
and MamH are involved in redox control of magnetite biomineralization in Magnetospirillum gryphis-
waldense. Mol Microbiol. 2013 Sep; 89(5):872–86. https://doi.org/10.1111/mmi.12317 PMID:
23889511
66. Mu¨ller FD, Raschdorf O, Nudelman H, Messerer M, Katzmann E, Plitzko JM, et al. The FtsZ-Like Pro-
tein FtsZm of Magnetospirillum gryphiswaldense Likely Interacts with Its Generic Homolog and Is
Required for Biomineralization under Nitrate Deprivation. J Bacteriol. 2014 Feb 1; 196(3):650–9.
https://doi.org/10.1128/JB.00804-13 PMID: 24272781
67. Ding Y, Li J, Liu J, Yang J, Jiang W, Tian J, et al. Deletion of the ftsZ-Like Gene Results in the Produc-
tion of Superparamagnetic Magnetite Magnetosomes in Magnetospirillum gryphiswaldense. J Bacter-
iol. 2010 Feb; 192(4):1097–105. https://doi.org/10.1128/JB.01292-09 PMID: 20023033
68. Li Y, Katzmann E, Borg S, Schu¨ler D. The Periplasmic Nitrate Reductase Nap Is Required for Anaero-
bic Growth and Involved in Redox Control of Magnetite Biomineralization in Magnetospirillum gryphis-
waldense. J Bacteriol. 2012 Sep 15; 194(18):4847–56. https://doi.org/10.1128/JB.00903-12 PMID:
22730130
69. Li Y, Raschdorf O, Silva KT, Schu¨ler D. The Terminal Oxidase cbb3 Functions in Redox Control of
Magnetite Biomineralization in Magnetospirillum gryphiswaldense. J Bacteriol. 2014 Jul; 196
(14):2552–62. https://doi.org/10.1128/JB.01652-14 PMID: 24794567
70. Li Y, Sabaty M, Borg S, Silva KT, Pignol D, Schu¨ler D. The oxygen sensor MgFnr controls magnetite
biomineralization by regulation of denitrification in Magnetospirillum gryphiswaldense. BMC Microbiol.
2014 Jun 10; 14(1):153.
71. Fillat MF. The FUR (ferric uptake regulator) superfamily: Diversity and versatility of key transcriptional
regulators. Arch Biochem Biophys. 2014 Mar 15; 546:41–52. https://doi.org/10.1016/j.abb.2014.01.
029 PMID: 24513162
72. Uebe R, Voigt B, Schweder T, Albrecht D, Katzmann E, Lang C, et al. Deletion of a fur-like gene affects
iron homeostasis and magnetosome formation in Magnetospirillum gryphiswaldense. J Bacteriol.
2010 Aug; 192(16):4192–204. https://doi.org/10.1128/JB.00319-10 PMID: 20562310
73. Rong C, Huang Y, Zhang W, Jiang W, Li Y, Li J. Ferrous iron transport protein B gene (feoB1) plays an
accessory role in magnetosome formation in Magnetospirillum gryphiswaldense strain MSR-1. Res
Microbiol. 2008 Oct; 159(7–8):530–6. https://doi.org/10.1016/j.resmic.2008.06.005 PMID: 18639631
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 17 / 19

74. Suzuki T, Okamura Y, Calugay RJ, Takeyama H, Matsunaga T. Global Gene Expression Analysis of
Iron-Inducible Genes in Magnetospirillum magneticum AMB-1. J Bacteriol. 2006 Mar; 188(6):2275–9.
https://doi.org/10.1128/JB.188.6.2275-2279.2006 PMID: 16513757
75. Taoka A, Umeyama C, Fukumori Y. Identification of iron transporters expressed in the magnetotactic
bacterium Magnetospirillum magnetotacticum. Curr Microbiol. 2009 Feb; 58(2):177–81. https://doi.
org/10.1007/s00284-008-9305-7 PMID: 18972162
76. Wang Q, Wang M, Wang X, Guan G, Li Y, Peng Y, et al. Iron Response Regulator Protein IrrB in Mag-
netospirillum gryphiswaldense MSR-1 Helps Control the Iron/Oxygen Balance, Oxidative Stress Toler-
ance, and Magnetosome Formation. Appl Environ Microbiol. 2015 Dec; 81(23):8044–53. https://doi.
org/10.1128/AEM.02585-15 PMID: 26386052
77. Lin W, Deng A, Wang Z, Li Y, Wen T, Wu L-F, et al. Genomic insights into the uncultured genus “Can-
didatus Magnetobacterium” in the phylum Nitrospirae. ISME J. 2014 Dec; 8(12):2463–77. https://doi.
org/10.1038/ismej.2014.94 PMID: 24914800
78. Sakaguchi T, Burgess JG, Matsunaga T. Magnetite formation by a sulphate-reducing bacterium.
Nature. 1993 Sep; 365(6441):47.
79. Sakaguchi T, Arakaki A, Matsunaga T. Desulfovibrio magneticus sp. nov., a novel sulfate-reducing
bacterium that produces intracellular single-domain-sized magnetite particles. Int J Syst Evol Micro-
biol. 2002 Jan; 52(1):215–21.
80. Grant CR, Rahn-Lee L, LeGault KN, Komeili A. Genome Editing Method for the Anaerobic Magneto-
tactic Bacterium Desulfovibrio magneticus RS-1. Appl Env Microbiol. 2018 Nov 15; 84(22):e01724–
18.
81. Byrne ME, Ball DA, Guerquin-Kern J-L, Rouiller I, Wu T-D, Downing KH, et al. Desulfovibrio magneti-
cus RS-1 contains an iron- and phosphorus-rich organelle distinct from its bullet-shaped magneto-
somes. Proc Natl Acad Sci. 2010 Jul 6; 107(27):12263–8. https://doi.org/10.1073/pnas.1001290107
PMID: 20566879
82. Kolinko S, Richter M, Glo
¨ckner F-O, Brachmann A, Schu¨ler D. Single-cell genomics of uncultivated
deep-branching magnetotactic bacteria reveals a conserved set of magnetosome genes. Environ
Microbiol. 2016 Jan; 18(1):21–37. https://doi.org/10.1111/1462-2920.12907 PMID: 26060021
83. Descamps ECT, Monteil CL, Menguy N, Ginet N, Pignol D, Bazylinski DA, et al. Desulfamplus magne-
tovallimortis gen. nov., sp. nov., a magnetotactic bacterium from a brackish desert spring able to biomi-
neralize greigite and magnetite, that represents a novel lineage in the Desulfobacteraceae. Syst Appl
Microbiol. 2017 Jul; 40(5):280–9. https://doi.org/10.1016/j.syapm.2017.05.001 PMID: 28622795
84. Monteil CL, Perrière G, Menguy N, Ginet N, Alonso B, Waisbord N, et al. Genomic study of a novel
magnetotactic Alphaproteobacteria uncovers the multiple ancestry of magnetotaxis. Environ Microbiol.
2018 Dec; 20(12):4415–30. https://doi.org/10.1111/1462-2920.14364 PMID: 30043533
85. Lefèvre CT, Menguy N, Abreu F, Lins U, Po
´sfai M, Prozorov T, et al. A cultured greigite-producing
magnetotactic bacterium in a novel group of sulfate-reducing bacteria. Science. 2011 Dec 23; 334
(6063):1720–3. https://doi.org/10.1126/science.1212596 PMID: 22194580
86. DeLong EF, Frankel RB, Bazylinski DA. Multiple Evolutionary Origins of Magnetotaxis in Bacteria. Sci-
ence. 1993; 259(5096):803–6. https://doi.org/10.1126/science.259.5096.803 PMID: 17809345
87. Abreu F, Cantão ME, Nicola
´s MF, Barcellos FG, Morillo V, Almeida LG, et al. Common ancestry of
iron oxide- and iron-sulfide-based biomineralization in magnetotactic bacteria. ISME J. 2011 Oct; 5
(10):1634–40. https://doi.org/10.1038/ismej.2011.35 PMID: 21509043
88. Lefèvre CT, Trubitsyn D, Abreu F, Kolinko S, Almeida LGP de, Vasconcelos ATR de, et al. Monophy-
letic origin of magnetotaxis and the first magnetosomes. Environ Microbiol. 2013; 15(8):2267–74.
https://doi.org/10.1111/1462-2920.12097 PMID: 23438345
89. Yang W, Li R, Peng T, Zhang Y, Jiang W, Li Y, et al. mamO and mamE genes are essential for magne-
tosome crystal biomineralization in Magnetospirillum gryphiswaldense MSR-1. Res Microbiol. 2010
Oct; 161(8):701–5. https://doi.org/10.1016/j.resmic.2010.07.002 PMID: 20674739
90. Quinlan A, Murat D, Vali H, Komeili A. The HtrA/DegP family protease MamE is a bifunctional protein
with roles in magnetosome protein localization and magnetite biomineralization. Mol Microbiol. 2011
May; 80(4):1075–87. https://doi.org/10.1111/j.1365-2958.2011.07631.x PMID: 21414040
91. Hershey DM, Ren X, Melnyk RA, Browne PJ, Ozyamak E, Jones SR, et al. MamO Is a Repurposed
Serine Protease that Promotes Magnetite Biomineralization through Direct Transition Metal Binding in
Magnetotactic Bacteria. PLoS Biol. 2016 Mar; 14(3):e1002402. https://doi.org/10.1371/journal.pbio.
1002402 PMID: 26981620
92. Hershey DM, Browne PJ, Iavarone AT, Teyra J, Lee EH, Sidhu SS, et al. Magnetite Biomineralization
in Magnetospirillum magneticum Is Regulated by a Switch-like Behavior in the HtrA Protease MamE. J
Biol Chem. 2016 19; 291(34):17941–52. https://doi.org/10.1074/jbc.M116.731000 PMID: 27302060
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 18 / 19

93. Wetmore KM, Price MN, Waters RJ, Lamson JS, He J, Hoover CA, et al. Rapid Quantification of
Mutant Fitness in Diverse Bacteria by Sequencing Randomly Bar-Coded Transposons. mBio. 2015
Jul 1; 6(3):e00306–15. https://doi.org/10.1128/mBio.00306-15 PMID: 25968644
94. Chen H, Zhang S-D, Chen L, Cai Y, Zhang W-J, Song T, et al. Efficient Genome Editing of Magnetos-
pirillum magneticum AMB-1 by CRISPR-Cas9 System for Analyzing Magnetotactic Behavior. Front
Microbiol. 2018 Jul 17; 9.
95. Peters JM, Colavin A, Shi H, Czarny TL, Larson MH, Wong S, et al. A Comprehensive, CRISPR-
based Functional Analysis of Essential Genes in Bacteria. Cell. 2016 Jun 2; 165(6):1493–506.
96. Peters JM, Koo B-M, Patino R, Heussler GE, Hearne CC, Qu J, et al. Enabling genetic analysis of
diverse bacteria with Mobile-CRISPRi. Nat Microbiol. 2019 Feb; 4(2):244–50. https://doi.org/10.1038/
s41564-018-0327-z PMID: 30617347
97. Tay A, Pfeiffer D, Rowe K, Tannenbaum A, Popp F, Strangeway R, et al. High-Throughput Microfluidic
Sorting of Live Magnetotactic Bacteria. Appl Env Microbiol. 2018 Sep 1; 84(17):e01308–18.
98. Tay A, McCausland H, Komeili A, Carlo DD. Nano and Microtechnologies for the Study of Magnetotac-
tic Bacteria. Adv Funct Mater. 2019; 29(38):1904178.
99. Wen T, Guo F, Zhang Y, Tian J, Li Y, Li J, et al. A novel role for Crp in controlling magnetosome biosyn-
thesis in Magnetospirillum gryphiswaldense MSR-1. Sci Rep. 2016 Feb 16; 6:21156. https://doi.org/
10.1038/srep21156 PMID: 26879571
100. Lefèvre CT, Bennet M, Landau L, Vach P, Pignol D, Bazylinski DA, et al. Diversity of magneto-aerotac-
tic behaviors and oxygen sensing mechanisms in cultured magnetotactic bacteria. Biophys J. 2014 Jul
15; 107(2):527–38. https://doi.org/10.1016/j.bpj.2014.05.043 PMID: 25028894
101. Popp F, Armitage JP, Schu¨ler D. Polarity of bacterial magnetotaxis is controlled by aerotaxis through a
common sensory pathway. Nat Commun. 2014 Nov 14; 5:5398. https://doi.org/10.1038/ncomms6398
PMID: 25394370
102. Simmons SL, Sievert SM, Frankel RB, Bazylinski DA, Edwards KJ. Spatiotemporal distribution of
marine magnetotactic bacteria in a seasonally stratified coastal salt pond. Appl Environ Microbiol.
2004 Oct; 70(10):6230–9. https://doi.org/10.1128/AEM.70.10.6230-6239.2004 PMID: 15466570
103. Flies CB, Jonkers HM, de Beer D, Bosselmann K, Bo
¨ttcher ME, Schu¨ler D. Diversity and vertical distri-
bution of magnetotactic bacteria along chemical gradients in freshwater microcosms. FEMS Microbiol
Ecol. 2005 Apr 1; 52(2):185–95. https://doi.org/10.1016/j.femsec.2004.11.006 PMID: 16329905
104. Liu J, Zhang W, Du H, Leng X, Li J-H, Pan H, et al. Seasonal changes in the vertical distribution of two
types of multicellular magnetotactic prokaryotes in the sediment of Lake Yuehu, China. Environ Micro-
biol Rep. 2018; 10(4):475–84. https://doi.org/10.1111/1758-2229.12652 PMID: 29687636
105. Moisescu C, Ardelean II, Benning LG. The effect and role of environmental conditions on magneto-
some synthesis. Front Microbiol. 2014 Feb 11; 5.
106. Bazylinski DA, Lefèvre CT. Magnetotactic bacteria from extreme environments. Life Basel Switz. 2013
Mar 26; 3(2):295–307.
PLOS Genetics | https://doi.org/10.1371/journal.pgen.1008499 February 13, 2020 19 / 19