Skip to main content

Towards a 'chassis' for bacterial magnetosome biosynthesis: genome streamlining of Magnetospirillum gryphiswaldense by multiple deletions



Because of its tractability and straightforward cultivation, the magnetic bacterium Magnetospirillum gryphiswaldense has emerged as a model for the analysis of magnetosome biosynthesis and bioproduction. However, its future use as platform for synthetic biology and biotechnology will require methods for large-scale genome editing and streamlining.


We established an approach for combinatory genome reduction and generated a library of strains in which up to 16 regions including large gene clusters, mobile genetic elements and phage-related genes were sequentially removed, equivalent to ~ 227.6 kb and nearly 5.5% of the genome. Finally, the fragmented genomic magnetosome island was replaced by a compact cassette comprising all key magnetosome biosynthetic gene clusters. The prospective 'chassis' revealed wild type-like cell growth and magnetosome biosynthesis under optimal conditions, as well as slightly improved resilience and increased genetic stability.


We provide first proof-of-principle for the feasibility of multiple genome reduction and large-scale engineering of magnetotactic bacteria. The library of deletions will be valuable for turning M. gryphiswaldense into a microbial cell factory for synthetic biology and production of magnetic nanoparticles.


Magnetosomes are membrane-enclosed organelles that are synthesized by various aquatic bacteria for their magnetotactic navigation in the Earth’s geomagnetic field [1, 2]. Apart from their biological function as magnetic sensors, magnetosomes also represent microbially synthesized magnetic nanoparticles (MNP) consisting of monocrystalline magnetite (Fe3O4) or greigite (Fe3S4). Because of their strictly controlled biomineralization, bacterial magnetosomes have exceptional properties, such as high chemical purity and crystallinity, strong magnetization, and uniform sizes and shapes, which are largely unknown from chemically synthesized MNP [3,4,5]. This makes them highly attractive for a number of biotechnological and biomedical applications [6,7,8]. For examples, magnetosomes isolated from the magnetic bacterium Magnetospirillum gryphiswaldense were successfully tested as multimodal reporters for magnetic imaging [9, 10], nanocarriers for magnetic drug targeting [11,12,13], and for magnetic hyperthermia applications [14, 15]. In addition, the functionality of magnetosomes can be extended by genetically fusing foreign functional moieties and polypeptides, such as fluorophores, enzymes, antibodies, and organic shells [16,17,18,19,20,21,22] to magnetosome membrane anchors. Moreover, the bacteria were utilized as a model to study the molecular mechanisms of human diseases related to homologs of certain magnetosome proteins [23].

However, applications of bacteria and their magnetosomes so far have been hampered by the limited number of appropriate production strains and difficulties in their large-scale cultivation and genetic manipulation. One of the most extensively investigated model organisms for studying magnetosome biosynthesis is the freshwater alphaproteobacterium M. gryphiswaldense [24, 25]. It typically produces 15–25 cuboctahedral magnetite particles per cell that are about 40 nm in size [26], while genetic overexpression of gene clusters governing magnetosome biosynthesis generated an overproducing strain forming > 100 (up to 170) particles per cell with enlarged sizes [27]. Because of its genetic tractability and relatively straightforward cultivation, M. gryphiswaldense recently has also emerged as host strain for bioproduction and synthetic biology of magnetosomes [20, 21, 27,28,29,30,31].

Despite of this recent progress, since its isolation [24, 32, 33] the usability of the undomesticated M. gryphiswaldense as a biotechnological workhorse has been limited due to several unwanted features. For example, one obstacle is the rather fastidious and sometimes fluctuating growth, which makes cultivation difficult to reproduce at larger scale. In other bacteria, this erratic growth behavior has been attributed to the presence of prophage genes, which are often known to exhibit some latent activity, resulting in a negative impact on the robustness of growth and the performance of bioprocesses [34,35,36]. Another adverse feature is the inherent genetic instability of M. gryphiswaldense, and in particular of the magnetic phenotype, which makes genetic manipulation and magnetosome production cumbersome. For example, spontaneous loss or impairment of magnetosome biosynthesis has been observed frequently during subcultivation, which had been traced back to spontaneous deletions and rearrangements within the large genomic magnetosome island (MAI) [37,38,39]. This chromosomal region extends across about 100 kb and comprises discontiguous clusters of more than 30 genes responsible for magnetosome formation organized in the five polycistronic operons feoAB1, mms6, mamGFDC, mamAB and mamXY [1, 40]. In addition, the MAI harbors regions of irrelevant gene content and numerous mobile genetic elements, which might be responsible for the frequent rearrangements and loss of magnetic phenotype in M. gryphiswaldense [37, 38, 41].

For future synthetic biology applications as well as large-scale magnetosome bioproduction, a simplified and potentially more robust version of the M. gryphiswaldense genome would be highly beneficial. In other bacteria, moderate genome reduction, which comprises the targeted deletion of multiple dispensable genes, has been shown to optimize metabolic pathways, enhance the expression of recombinant protein productivity, and improve physiological performance and growth [42,43,44,45,46]. For instance, by removing non-essential, recombinogenic or mobile DNA and cryptic virulence genes, genome reduction of Escherichia coli resulted in several favorable properties, such as increased electroporation efficiency and improved propagation of recombinant genes [47]. Deletion of prophage genes improved growth and transformation efficiency in Corynebacterium glutamicum [34], enhanced genotypic stability in Pseudomonas putida [35, 48, 49], and increased robustness toward stress in Vibrio natriegens or Shewanella oneidensis MR-1 [36, 50]. Furthermore, deletion of active mobile genetic elements caused enhanced protein productivity in C. glutamicum [51], and increased transformability and reduced mutation rates in Acinetobacter baylyi [52].

In magnetotactic bacteria, comparable genome reduction approaches so far have been impeded because of the unavailability of efficient methods for large-scale engineering of these recalcitrant microorganisms. To overcome these current limitations, we recently started a systematic approach to engineer the model strain M. gryphiswaldense at the genome level. In a previous study, we established a method for large-scale deletion mutagenesis and utilized it for the generation of 24 single deletions covering about 167 kb of non-redundant genome content. We thereby identified large regions inside and outside the MAI irrelevant for magnetosome biosynthesis [53]. Here, we continued our work by constructing genome-reduced strains of M. gryphiswaldense with multiple combinatorial deletions of irrelevant and detrimental gene content. We provide a proof of concept for large-scale genome editing and improvement towards a future chassis [54], which may turn M. gryphiswaldense into a microbial cell factory for the synthetic biology and high-yield production of magnetic nanoparticles.


Overview over the experimental strategy

The features of a genome reduced future 'chassis' should include first the elimination of problematic and harmful gene content such as prophage genes as well as active IS elements known to cause genetic instability [50,51,52]. Second, the genome should be freed of as much of gene content unnecessary for magnetosome biosynthesis, growth and fitness under lab conditions as possible. Third, neutral and favorable scarless deletions should be combined into one or few single strains. Ultimately, the native biosynthetic gene clusters within the MAI plus multiple large portions of 'junk' between and adjacent to them should be substituted by a compact cassette comprising all key genes for magnetosome biosynthesis (plasmid pMDJM3). Final strains were tested for growth, fitness, and genetic stability (for an overview over the experimental workflow see Fig. 1a).

Fig. 1
figure 1

Overview over the experimental workflow (a), and the genomic positions of deletion targets in M. gryphiswaldense (b). Yellow circle (b) shows genes or gene sets targeted for multiple deletions. Grey: magnetosome island (MAI); black arrows: parts of predicted prophage sets and phage-related integrase and excisionase genes; green arrows: insertion element genes identified as most active in this study; purple: irrelevant gene clusters. Pink circle (b) indicates predicted prophage sets (dark red) and mobile genetic elements (light blue). The enlarged area shows the genetic organization of the native MAI with all five known magnetosome biosynthesis operons (red), genes of known or unknown function irrelevant for magnetosome formation (black), mobile genetic elements (light blue). The grey bars indicate the extent of regions M04 (~ 65 kb) and M13 (~ 100 kb), which were shown to be deletable en bloc in our previous study [53]. The presence of key magnetosome biosynthesis genes within the compact expression cassette of the applied vector pMDJM3 is indicated by red bars while connecting lines designate eliminated gene content

Identification and elimination of prophage genes

Since its isolation [24, 32, 33], our lab persistently experienced occasional problems with the cultivation of M. gryphiswaldense, such as poorly reproducible and fluctuating growth, which could not only be explained by unintended subtle variations in handling, media constituents and incubation alone. In other bacteria, similar observations could be traced back to the latent activity and induction of prophages, which are known to often have a negative impact on robustness of growth and the performance of bioprocesses [34]. In the genome of M. gryphiswaldense, we detected seven putative prophage regions (referred to as P1–P7) (Fig. 1b) by the phage search tool PHAST [55]. Two of them (P1, 26 kb, and P6, 34.1 kb) were predicted as intact (Figs. 1b and 2a, b), but upon closer inspection only P6 is contiguous and seems to have a full complement of typical phage genes (e.g. integrases, tail, and major capsid proteins), whereas P1 is interspersed with tRNA genes (Fig. 2a, b). Predicted prophage P2 seems to be incomplete as well, consistent with its small size (7.2 kb) and its accumulation of several transposon genes. Region P3 (20 kb) comprises some putative essential genes (e.g. encoding transcriptional regulators, chaperones) and a phage integrase intA1. Incomplete P4 (31.2 kb) and P5 (14.3 kb) are also interspersed with genes of unrelated, but important functions, e.g. chaperones of DnaJ-class, a transcriptional regulator and a ribonuclease, respectively. Incomplete P7 (11.3 kb) resides inside the part of the MAI (Fig. 1b) that was non-deletable in previous experiments [53]. From the identified putative phages, the following regions were selected as targets for deletion: P1 was divided into two parts excluding the essential tRNA genes (P1.2, 19.4 kb, and P1.3, 12.8 kb) that were both deleted separately (Fig. 2a). Since deletion of whole P6 failed, only genes encoding putative capsid proteins (7.65 kb) and a recombinase (hin2, 1.34 kb) were deleted separately (Fig. 2b). Furthermore, inside and adjacent to these predicted prophages, we identified several putative integrases and excisionases that might be involved in the reactivation of lysogenic prophages to the lytic cycle and decided to delete several candidates as well (intA1 1.28 kb, intA2 1.22 kb and alpA 225 bp) (Fig. 1b).

Fig. 2
figure 2

Molecular organization and analysis of predicted prophages P1 (a), P6 (b), deleted genes and growth characteristics of selected cultures (c, d). a, b Black: hypothetical genes or genes with known function; red: putative prophage genes; green: mobile genetic elements; blue: deleted genes. c Growth characteristics of ∆intA2 under aerobic and anaerobic conditions compared to WT. Growth of other prophage deletants (not shown) was virtually identical. d Growth profiles of mitomycin C-treated (MMC) cultures of a selected single prophage deletion mutant and the WT. Cultures were incubated for 8 h at MMC concentrations of 0.2 or 0.3 µg/ml, washed twice, and adjusted to the initial OD. Growth experiments were performed at 28 °C under aerobic conditions. Each strain was analyzed in triplicates, the curve represents the calculated average (standard deviations were < 5%)

Under aerobic and anaerobic conditions, growth of all prophage deletants was largely indistinguishable from the WT (data not shown). After incubation with 0.2 µg/ml MMC, which is known to trigger the cellular SOS response and to induce prophages to enter the lytic cycle [36, 56], growth was indistinguishable from the WT for most deletants. A notable exception was ∆hin2 (msr1_37790), which proved to be less sensitive and could be re-grown after incubation with up to 0.3 µg/ml MMC (Fig. 2c).

Deletion of active mobile genetic elements

Previous observations had revealed a genetic instability of the M. gryphiswaldense WT strain: first, the ability to form magnetosomes often became spontaneously reduced or lost entirely, which had been hypothesized to be caused by the presence and activity of numerous mobile elements, resulting in insertions by transposition activity [53], plus deletions and rearrangements caused by homologous recombination between identical copies [37, 38, 41]. Second, during the course of routine genetic manipulation, we frequently also observed spontaneous inactivation of introduced foreign genes, such as chromogenic reporters (e.g. gusA, unpublished observations) or genetic markers for antibiotic or counterselection (e.g. galK) [53]. Our preliminary analysis revealed that inactivation was often due to insertion of mobile genetic elements belonging to two types, each with two variants: the first type is a bipartite insertion element (in the following referred to as ISMgr2; Fig. 3a), composed of genes encoding a putative IS2 repressor TnpA, and an IS2 transposase TnpB, respectively. ISMgr2 belongs to the IS3 family that is common in many α-Proteobacteria [57]. Three copies of ISMgr2 (ISMgr2-1, ISMgr2-2 and ISMgr2-3) with 99.8% protein identity (99.9% nucleotide identity) are present in the genome of M. gryphiswaldense (Fig. 3a), with one of them residing within the MAI (Fig. 1b, ISMgr2-3). Two additional homologs of tnpB, termed ISMgr2-tnpB-hyp-1 and ISMgr2-tnpB-hyp-2, with lower (20.8%) protein identity (52.6% nucleotide identity) compared to the first three copies ISMgr2-1, ISMgr2-2 and ISMgr2-3 (Fig. 3b) could be identified. Each of these latter two homologs are associated with two conserved hypothetical genes upstream of the IS2 transposon gene tnpB instead of tnpA. We first deleted each of the five homologs individually in the WT background (1–1.25 kb each). As expected, single deletion mutants of all five strains displayed WT-like growth and magnetosome biosynthesis under aerobic and anaerobic conditions, and under oxidative and moderate heat stress (data not shown). We therefore later decided to sequentially delete these five homologs altogether (see below strains ∆TZ-05, ∆TZ-06, ∆TZ-09, ∆TZ-10 and ∆TZ-11 in Fig. 4).

Fig. 3
figure 3

Overview over identified active mobile genetic elements in M. gryphiswaldense. a One group of active insertion elements (ISMgr2tnpA and ISMgr2tnpB, green) belongs to the IS3 family, all three copies were deleted in this study. b The second variant of this group contains two hypothetical genes (black) and an ISMgr2tnpB gene. Both copies were deleted. c A further group of active mobile elements is represented by a transposon tandem (tn-tandem). The first transposon (tn1) of the tn-tandem is a member of the putative transposase of IS4/5 family and contains a DUF4096 domain known to bind the end of a transposon and to catalyze the movement of the transposon to another part of the genome by cut and paste or replicative transposition mechanism. The second tn-tandem transposon (tn2) contains a DDE domain, named after a conserved amino acid triad Asp, Asp, Glu, the active site [57]. d tn2 is present in two single copies in the WT genome. Two copies of tn-tandem were deleted in ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA

Fig. 4
figure 4

Scheme of multiple deletions and their genotypes. Black arrows indicate successful deletions while red bars show failed deletions. Colored letters indicate final strains with ∆mamAB (blue) or ∆M13 (green) deletions. Extents of genome reduction of multiple deletion mutants are given in bp and % of the WT genome

The second type of active mobile elements is represented by a transposon tandem (tn-tandem) of genes sharing 52.3% nucleotide identity (31.25% protein identity) (Fig. 3c): its first gene (tn1) encodes a putative transposase of the IS4/5 family [57], and the second gene (tn2) a DDE domain transposase [57] (Fig. 3c). This tandem pair is present in the genome of M. gryphiswaldense in 19 identical (100% nt) copies, and in addition tn2 alone in two more identical single copies. Four of the 19 tn-tandem pairs reside within the MAI, with two of them comprised within the deletable regions M04 and M13 [53] (Fig. 1b). However, due to the unavailability of multiplex genetic tools for M. gryphiswaldense, the high copy number of the tn-tandem pairs proved prohibitive for sequential or simultaneous deletion of all copies.

Identification and elimination of further gene clusters irrelevant for cell growth and magnetosome biosynthesis

Next, to eliminate further larger non-essential chromosomal stretches outside the MAI, we exemplary targeted two gene clusters that are likely irrelevant for magnetosome biosynthesis and fitness in lab conditions (Additional file 1: Table S1): (i) A nif operon comprising 16 genes, namely nifWABZTHDK, fixABC, draGT and three ferredoxin genes (msr1_18560; msr1_18600; msr1_18640). The nif operon is likely linked to nitrogen fixation in M. gryphiswaldense [58] (Additional file 1: Fig. S1) which is irrelevant under the denitrifying conditions optimal for magnetosome biosynthesis [28]. We generated a mutant in which ~ 20 kb of this nif cluster comprising 20 genes were deleted (Additional file 1: Fig. S1). (ii) Several uncharacterized clusters encoding a putative non-ribosomal peptide synthetase (NRPS) and a polyketide synthase (PKS) were predicted [59]. Since it was unlikely to be necessary for magnetosome biosynthesis, we deleted three large ORFs encoding putative PKS proteins from one of the clusters (termed pks) extending over ~ 40 kb (Fig. 1b; Additional file 1: Table S1). As expected, strains harboring single deletions in nif or pks clusters were indistinguishable from the WT with regard to magnetosome formation, cell growth and motility, which confirmed their irrelevance for magnetosome biosynthesis and fitness under lab conditions (Additional file 1: Table S1, Fig. S4).

Combinatory mutagenesis

In order to combine all previously tested favorable or neutral deletions into one or two single strains, we employed the following strategy (Fig. 4): starting with the ∆intA2 strain as a parent, we first proceeded by deleting further selected prophage genes (∆TZ-01–∆TZ-04), then continued with the mobile genetic elements (∆TZ-05 and ∆TZ-06) and further prophage genes and IS elements (∆TZ-07–∆TZ-11) and ended with deletion of irrelevant gene clusters and magnetosome biosynthesis genes (∆TZ-12–∆TZ-17). One round of deletion was completed as soon as the loss of kanamycin resistance marker (Kmr) was verified by replica plating. After each round, magnetic responses of mutant strains as well as their growth under aerobic and anaerobic conditions, oxidative and moderate heat stress were tested. This was found to be WT-like for all offspring strains including ∆TZ-15 (Additional file 1: Fig. S2).

After fifteen successful rounds of deletions, the resulting mutant ∆TZ-15 was used as a parent to delete large parts of the MAI including all magnetosome biosynthesis operons. In a previous study, a contiguous stretch of ~ 66 kb termed region M04 was found to be deletable in the WT-background, including all mam and mms6 operons (~ 27 kb) plus ~ 39 kb of irrelevant or problematic gene content, such as two copies of tn-tandem (Fig. 1b, enlargement, grey bar) [53]. This had no obvious effects on growth, and magnetosome biosynthesis could be restored to WT-level by complementation with a compact expression cassette comprising the mam and mms6 operons only [53]. In addition, a ~ 100 kb region termed M13 (Fig. 1b) could be excised, again including all mam and mms6 operons, plus an additional ~ 33 kb flanking region. Despite of its slightly impaired growth in oxidative stress conditions [53], the M13 region was chosen as an additional target to generate a strain with the largest possible genome reduction.

However, we failed to delete M04 in strain ∆TZ-15 despite of several attempts (Fig. 4), although its deletion had been readily achieved before in the WT background [53]. Instead, upon repeated attempts of conjugation and counterselection, we obtained a number of conspicuous clones with either magnetic or non-magnetic phenotypes, which had supposedly excised the deletion target as suggested by PCR, but lost their insensitivity against kanamycin, indicating that parts of the suicide vector harboring the Kmr marker were likely still maintained in the genome. In our previous study, similar observations could be traced back to the inactivation of the galK gene encoding the lethal galactokinase, followed by spontaneous rearrangements in the absence of rigorous counterselection [53]. This explained our failure to enforce the proper deletion during counterselection in the presence of galactose, and in fact, the entire M04 region was still present in the genome (see Additional file 1: Fig. S3 for detail). To circumvent this problem, we separately deleted the essential mamABop first in strain ∆TZ-15, yielding strain ∆TZ-16, in which we attempted subsequent deletion of the residual M04 region. Several kanamycin sensitive (Kms), non-magnetic clones were obtained in this regime, which however again yielded diverse PCR products only roughly similar to the expected size spanning over the targeted excision site. Nevertheless, one of the clones (still tentatively termed ∆TZ-16), was selected as parent for later re-insertion of pMDJM3 harboring a compact version of the magnetosome biosynthesis gene clusters (see below).

In contrast to the troublesome M04 deletion, one-step deletion of the even larger region M13 in the background of ∆TZ-15 was obtained readily and yielded plenty of expected non-magnetic clones, in which the proper deletion of M13 could be confirmed by PCR spanning over the targeted excision site. This yielded strain ∆TZ-17. Like the respective single deletion mutants ∆M04 and ∆M13 in the WT background [53], both intermediate strains ∆TZ-16 and ∆TZ-17 showed conspicuous irregularly shaped electron dense particles (EDPs) between 10 and 125 nm in size in electron micrographs (see Fig. 6a below, white arrows), which were previously shown to be rich in potassium, phosphorus and oxygen, and to be unrelated to magnetosome biosynthesis [53].

In the final step, restoration of magnetosome biosynthesis was attempted in the two multiple deletion strains. This was achieved by insertion of pMDJM3 or pMDJM3-gusA, variants of pTpsMAG1 [60] harboring the compact set of mam/mms/feo genes and lox sites for restoration of antibiotic resistance to generate a marker-less mutant, and in case of pMDJM3-gusA in addition encoding the enzyme GusA (glucuronidase) as a chromogenic reporter. The gusA gene was added next to the mamXYop as entrapment for spontaneous mutations in a genetic stability assay to ∆TZ-16 and ∆TZ-17 (see below). As control, gusA was also inserted into the WT strain at the same genomic position next to the mamXYop as in pMDJM3. The region downstream of mamXYop was chosen as site for gusA insertion, since spontaneous deletions, insertions and rearrangements of this particular region were observed repeatedly as a virtual hotspot during routine genetic manipulation (unpublished observations). This is possibly caused by its close proximity (~ 11.4 kb) to the two tn-tandem copies described above, and often accompanied by impaired magnetosome phenotypes akin a mamXYop deletion [61]. Thus, the strains ∆TZ-16::MAG-gusA, ∆TZ-17::MAG, ∆TZ-17::MAG-gusA and WT-gusA (Fig. 4) were generated.

Genome analysis of final multiple mutant strains

To verify the multiple introduced deletions, as well as possible unintended mutations and rearrangements that might have occurred during the numerous rounds of manipulation, at this point the two final multiple mutant strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG were subjected to genome resequencing. In strain ∆TZ-16::MAG-gusA this revealed that the region M04, which we attempted to delete in the last step, was still present as already suspected, except for mamABop, which had been removed already in the previous step in ∆TZ-16. As a consequence, ∆TZ-16::MAG-gusA is merodiploid for all magnetosome operons but mamABop and feoAB1op. All introduced mam/mms/feo genes were found to be present next to endogenous mamXYop (Fig. 5a, red box), although some with silent or neutral point mutations. However, conspicuously, the order of the introduced operons (feoAB1-mamAB-mamGFDC-mms6-gusA-mamXY) was shuffled compared to their original order on pMDJM3-gusA (mamAB-feoAB1-mamXY-gusA-mms6-mamGFDC). Apart from the failed M04 deletion, all other deletions introduced into ∆TZ-16::MAG-gusA were exactly as intended. However, besides a number of point mutations, a few larger indel mutations were found in genome regions likely to be irrelevant for magnetosome biosynthesis. These include the pORFM-GalK-M04 suicide vector within msr1_03120 (nt position 305,858), a 178 bp spontaneous deletion at nt position 2,599,005 in msr1_24320 (encoding a filamentous hemagglutinin) and an insertion of a copy of ISMgr2 at nt position 3,961,873 (with msr1_37870 encoding a phytochrome-like protein).

Fig. 5
figure 5

Schematic presentation showing an overview of ∆TZ-16::MAG-gusA (a) and ∆TZ-17::MAG (b) genotypes. Green circle shows the final genotype including unexpected insertions or deletions by resequencing of strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG. Grey: MAI; white: M13 or mamABop deletions; black arrows: parts of predicted prophage sets, integrase and excisionase genes; green: most active insertion element genes; purple: irrelevant gene cluster. Enlargements indicate unexpected duplication of vector remnants. In ∆TZ-16::MAG-gusA (a) remnants and duplications of up- and downstream regions of pORFM-GalK-M04 suicide vector are still located within msr1_03120 (encoding a putative secreted effector protein PipB) at the position (305,858 nt) targeted for deletion. As observed before, parts of up- and downstream homologous regions were found fragmented and duplicated. pMDJM3-gusA of strain ∆TZ-16::MAG-gusA has been inserted within the intergenic region between mamY (msr1_03880) and the adjacent transposon gene (msr1_03890) (red box). b pMDJM3 in strain ∆TZ-17::MAG is located at nt position 699,709 (red box) within the ruvB gene (msr1_07040) encoding a putative holliday junction ATP-dependent DNA helicase. Remnants and duplication of pORFM-GalK-pks suicide vector are located at nt position 1,646,447 (intergenic region upstream of msr1_15660) in strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG. Additionally, an unintended spontaneous 178 bp deletion in fhaB 2 gene (msr1_24320) encoding a filamentous hemagglutinin is present at nt position 2,599,005 (green circle, orange), and an insertion of a copy of ISMgr2 at nt position 3,961,873 into cph1 40 gene (msr1_37870) encoding a phytochrome-like protein was found in both strains (green circle, dark green)

In ∆TZ-17::MAG the entire M13 region was confirmed to be deleted exactly as intended (Fig. 5b, green circle). pMDJM3 was inserted at nt position 699,709 (Fig. 5b, green circle, red box). All introduced mam/mms/feo genes were found to be identical to pMDJM3 with respect to sequences and order. In addition to the successful M13 deletion, also all other introduced deletions were exactly as intended. A short remnant (3 558 bp) of suicide vector pORFM-GalK-pks was found inserted at nt position 1,646,447 (considered to be neutral), showing a duplication of the downstream homologous region like in ∆TZ-16::MAG-gusA. Again, the same spontaneous indel mutations in other chromosomal regions as in strain ∆TZ-16::MAG-gusA were also present in ∆TZ-17::MAG, indicating that these mutations had occurred already at an earlier stage of mutagenesis.

Phenotypic characterization of ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA

Growth characteristics and magnetosome biomineralization

Complementation of non-magnetic ∆TZ-16 (lacking 3.489% of the WT genome) and ∆TZ-17 (lacking 5.477%) with pMDJM3 restored the formation of WT-like magnetosome numbers and sizes, and cells had electron dense particles (EDP) (Fig. 6a, white arrows), similar as observed before in the corresponding single deletion strains of the eliminated parts of the MAI [53]. Microplate-scale experiments with strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA under aerobic and anaerobic nitrate-reducing conditions indicated WT-like or slightly delayed cell growth compared to WT (Fig. 6b). To analyze growth at higher cell densities, strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA were in addition cultivated in a larger volume (3 l) in a bioreactor under controlled anaerobic conditions, which are known to be optimal for magnetosome biomineralization [28, 62]. Figure 6c shows exemplary results for strain ∆TZ-16::MAG-gusA compared to the WT. Both strains reached a final OD of > 0.5, compared to only ca. 0.1 typically observed in microplate growth. Again, growth of strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA was WT-like, indicating that loss of the eliminated genes was neutral for growth under controlled conditions. Strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG were also tested regarding their growth performance after challenging them with the antibiotic MMC. Similar as the single deletion strain of the putative phage integrase gene hin2, both strains survived concentrations up to 0.3 µg/ml MMC, while WT was entirely inhibited at 0.3 µg/ml MMC (Additional file 1: Fig. S4).

Fig. 6
figure 6

Phenotypic characterization of multiple deletion mutants. a Electron micrographs of the non-magnetic pre-'chassis' strains ∆TZ-16 and ∆TZ-17 and the final complemented prospective chassis ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA (scale bars 500 nm), white arrows indicate EDPs and b cell growth under aerobic and anaerobic conditions as well as oxidative stress (H2O2) and heat stress (33 °C). Each strain was analyzed in triplicates and the curves show the average while standard deviation was below 5%. c Growth curves and Cmag (i.e., a proxy for the average magnetic orientation of bacterial cells in liquid media based on light-scattering [63]) of WT strain and ∆TZ-16::MAG-gusA during anaerobic fermentation

Stability of the reporter gene gusA and the magnetic phenotype

To analyze whether the combined multiple deletion of IS elements in M. gryphiswaldense affects the incidence of spontaneous mutations, we employed an assay to estimate the genetic stability of expressed foreign genes as well as the stability of the magnetic phenotype using the reporter gene gusA as a 'trap' (Additional file 1: Fig. S5A), similar as reported for other marker genes in different bacteria [64]. Mutational inactivation of gusA causes the loss of the ability to cleave X-Gluc into blue dye, hence resulting in white (magnetosome-free) or brownish (magnetosome forming) colonies (Additional file 1: Fig. S5). After ten sequential passages under aerobic conditions (alternating between 4 h at 4 °C and 44 h at 28 °C, conditions which were previously found to favor spontaneous mutations [41], 12 independent parallels (equivalent to ~ 4.2 * 103 cells for each strain and time point) were plated and visually screened. Out of ~ 2.5 * 104 cells in total, 2–3% of colonies had lost their blue color. Overall, among the about 7.1 * 102 white clones, we analyzed 192 white clones of each WT-gusA, ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA in which gusA was found to be inactivated by different types of mutations, including point mutations (80.6%), insertions (19.1%) and deletions (0.3%). The majority of the point mutations observed were base deletions (40%) and insertions (50%) causing frame shifts, while base substitutions represented the minority (10%). The types of point mutation were independent from time point or strain, and most mutations were found within a range of 500 bp of the gusA gene encoding the catalytic center of the GusA enzyme [65]. Furthermore, deletions of 62 bp of the gusA gene were found within the 25% N-terminal portion of the GusA protein.

All larger gusA insertions were found to belong exclusively to the two types of IS elements that we had already identified in our preliminary experiments described above [ISMgr2 (7.3%) and tn-tandem (92.7%)], present both in cells before (t0) and after (t10) passaging. The high frequency observed for tn-tandem insertions might have been probably caused by its high abundance or close proximity (~ 11.4 kb) of a copy to the gusA reporter. While the total number of mutations between time points t0 and t10 did not significantly vary (Table 1), the number of all insertions was substantially reduced by ca. 60–75% in ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA compared to WT-gusA, likely due to the successful elimination of several active IS elements described above.

Table 1 Overview of identified mutations in the reporter gene gusA

In addition to mutations in gusA, reduction or loss of the magnetic phenotype was found in a minority of clones from white or brown colonies which had lost their blue color. Reduced magnetic phenotypes displaying WT-like magnetite crystals flanked by flake-like particles could be observed in several of WT-gusA clones after ten passages (Additional file 1: Fig. S5B), which likely indicates a second mutation (i.e., in addition to the point mutations within gusA) in mamXYop [61]. Furthermore, the loss of the magnetic phenotype coincident with gusA inactivation could be observed in three clones of ∆TZ-16::MAG-gusA for time point t10. This could be the result of spontaneous homologous recombination between identical stretches of DNA in this partially merodiploidic strain.


In this study, we established an approach for large-scale combinatory genome reduction of the magnetotactic bacterium M. gryphiswaldense. By repeated circles of deletion, we generated a library of strains in which different multiple genomic segments were erased. These strains might each serve as different starting points in future genome streamlining approaches by recombination with further favorable deletions and insertions.

In total, we completed the combination of 16 single deletions from this and our previous work [53] into each of the two strains ∆TZ-16 and ∆TZ-17, in which in addition large parts of the fragmented MAI were functionally replaced by a compact version of the magnetosome biosynthesis gene clusters.

On average, one round of deletion typically took about 3 weeks, and after some technical streamlining, the 16 subsequent rounds could have been completed in about 12 months of work. Independent of the target size, all rounds of successful deletions were largely completed with similar efficiencies as for respective single deletions. An exception was the unsuccessful deletion of the M04 region which might be especially problematic due to the abundance of transposon genes close to the regions targeted for homologous recombination. Several undesired mutations and spontaneous rearrangements were found to have occurred during recursive deletions. This emphasizes the need of genome resequencing of key intermediates and final strains. Again, most of these spontaneous rearrangements were caused by either homologous or illegitimate recombination and could be traced back to the spontaneous inactivation of the lethal galK marker harbored on the suicide vectors for homologous recombination, thereby preventing effective counterselection of proper double-crossovers.

However, despite of these caveats, we succeeded in the construction of one final strain (∆TZ-17) with a genome reduction by non-overlapping ~ 227,600 bp, which is equivalent to about 5.5% of the entire genome. In this strain all targeted deletions and the reinsertion of the compacted magnetosome gene clusters were found to be exactly as intended, with only few minor spontaneous mutations in regions irrelevant for growth and magnetosome biosynthesis under laboratory conditions. This confirms that if used with caution, the method is sufficiently efficient and reliable for multiple genome editing.

In other bacteria, multiple genome reductions of different extents by similar approaches were previously reported. For example, one of the first studies in E. coli K12 resulted in a genome reduction up to 15% [47], and the genome of E. coli could be further shrunk by > 29%, changing cell size and nucleoid organization of engineered cells [66]. A “MiniBacillus” was constructed from Bacillus subtilis, in which a total of 42.3% was eliminated [67]. However, top-down genome reduction approaches in model organisms other than E. coli and B. subtilis have been more limited, and in some cases involved the combined, stepwise efforts of several labs [43]. For example, in C. glutamicum multiple approaches resulted first in the targeted deletion of 11 distinct regions with a total size of 250 kilobase pair (kbp) [68, 69], followed by an untargeted approach via insertion and excision [70]. In another random approach, 42 mutants in a range of 0.2–186 kb were generated, which revealed a total of 393.6 kb (11.9%) of the C. glutamicum R genome to be non-essential under standard laboratory conditions [70]. More recently, five of the 36 single large deletions identified by Unthan et al. [46] were later combined in a chassis strain of C. glutamicum, in which 13.4% of the genome were eliminated [45]. Similar approaches with P. putida resulted in genome reductions of 4.1% [71], 4.3% [48] and 4.12% [72], respectively, as well as in Lactococcus lactis (2.83% reduction) [73], Streptomyces avermitilis (1.4 Mb reduction) [74, 75], while in S. chattanoogensis 1.3 Mb and 0.7 Mb regions were eliminated [76]. With a comparable reduction by nearly 5.5% of the genome, our study represents the first proof-of-principle for the feasibility of similar targeted approaches in a magnetotactic bacterium.

While many of the genome streamlining approaches described above led to beneficial properties, such as improved growth and recombinant protein production, as well as robustness against several stresses [34, 45, 47, 50, 77], others resulted in negative effects, such as growth deficiencies, decreased resistance against antibiotics and under several stress conditions, and reduced transformation efficiencies [45, 78]. In our study, most deletions were neutral with respect to magnetosome biosynthesis and growth. For example, our preliminary analysis suggested that neither the conjugation efficiency with replicative and insertional plasmids, nor the weak latent propensity of spontaneous cell lysis was affected in strains ∆TZ-16::MAG-gusA and ∆TZ-17::MAG-gusA (data not shown). As observed in our previous study, deletion of mamABop in ∆TZ-16 and ∆TZ-17 resulted in a growth advantage which became lost after 're-magnetization' by complementation. Among the several putative prophage genes of M. gryphiswaldense, only deletion of the recombinase gene hin2 from P6 had an effect and resulted in a slightly improved resilience to mitomycin c (MMC)-induced stress in the final strains. This provides an indirect hint that P6 may be an active prophage, whose excision might be induced by MMC in WT cells, but further work, such as the identification of phage particles, will be necessary to confirm this. However, the combined deletion of other phage-related genes did neither further increase MMC resistance, nor generally enhance growth. On the contrary, deletion of the M13 region and prophage genes slightly impaired growth in the presence of oxidative stress, possibly for similar reasons as suggested by Wang et al. [78], who found that the presence of cryptic prophages may help bacteria to cope with adverse conditions and provide multiple benefits. For comparison, ∆TZ-16 lacks the ∆M13 deletion and therefore could be useful as an alternative parent strain with improved growth characteristics in follow-up genome streamlining studies.

From the ~ 120 transposable elements predicted in the 4.155 Mbp genome of M. gryphiswaldense, 30 are encoded within the ca. 100 kb MAI, and nine in addition in its ~ 33 kb adjacent region [1, 53]. However, our systematic approach revealed that only a minority of them, belonging to two families, seems to be responsible for the majority of spontaneous insertions. We detected an increased stability of the reporter gusA in both final multiple deletion strains, which was likely a result of the successful elimination of all ISMgr2 elements described above, including one from the MAI. Future approaches should also aim for the removal of ISMgr2-1, ISMgr2-2 and ISMgr2-3 which might further decrease the rate of spontaneous mutations. However, deletion of multiple copies of the tn-tandem, the second group of identified troublemakers, or even the generation of a chassis stripped of all copies as accomplished in several other bacteria [51, 52] is currently not within realistic reach, due to the numerous abundance and extensive sequence similarity between multiple copies of IS elements, as well as their persistent tendency to spread during genetic manipulation.


Overall, in this study we succeeded in further domestication and large-scale engineering of magnetotactic bacteria and showed the potential of combining multiple scarless deletions with high precision. We also generated a library of deletions, which represent building blocks for recombination with favorable deletions and insertions that can be used for the construction of improved 'chassis' strains in the future. Ultimately, this may turn M. gryphiswaldense into a versatile platform and microbial cell factory for synthetic biology and magnetosome production.


Bacterial strains, vectors, and cultivation conditions

Bacterial strains and plasmids used in this work are listed in Additional file 1: Table S1. E. coli strains were grown as previously reported [79]. E. coli WM3064 strains were grown in lysogeny broth (LB) medium supplemented with 25 µg/ml (final concentration) kanamycin (Km), and 1 mM dl-α,ε-diaminopimelic acid (DAP) at 37 °C. Liquid cultures of M. gryphiswaldense strains were grown microaerobically in flask standard medium (FSM) [28] at 28 °C under moderate shaking (120 rpm). Strains carrying the suicide or complementation plasmid were cultivated by adding 5 µg/ml Km. For cultivation on solid LB medium or FSM, 1.5% (w/v) agar was added. Cultivation from single M. gryphiswaldense colonies was conducted by transferring cell material into 150 µl FSM in 96-deep-well-plates (Eppendorf, Hamburg, Germany), prior to gradually increasing the culture volume. Optical density (OD) and magnetic response (Cmag, i.e., a proxy for the average magnetic orientation of bacterial cells in liquid media based on light-scattering) of cells in the exponential growth phase were measured photometrically at 565 nm as previously described [63].

Growth experiments of M. gryphiswaldense were performed by using pre-cultures grown for two daily passages under microaerobic conditions at 28 °C. Cultures were adjusted to an initial OD of 0.01 and grown in a microplate reader (Tecan) under aerobic conditions at 28 °C or moderate heat stress at 33 °C. For induction of oxidative stress, 20 µM H2O2 were added prior to starting the growth experiments.

For cell growth after mitomycin C (MMC) induction, pre-cultures were adjusted to an initial OD of 0.08 and treated with MMC concentrations of 0.1–0.3 µg/ml for 8 h. Non-induced strains (0 µg/ml) served as controls. Then cultures were washed twice in FSM, and an initial OD of 0.01 was used to start growth experiments in the microplate reader under aerobic conditions at 28 °C.

In preliminary experiments, conditions could be defined (i.e. incubation with 0.2 µg/ml MMC for 8 h) in which growth of the WT was already somewhat impaired, yet still reached substantial yields (final OD of ca. 0.3), while slightly increased MMC concentrations (0.3 µg/ml) entirely abolished growth. Therefore, we used 8 h and 0.2–0.3 µg/ml MMC as efficient incubation conditions to analyze survival of mutants compared to the WT strain.

For cultivation in the fermenter, modified FSM was used adding 10 mM NaNO3 instead of 4 mM NaNO3 as alternative electron acceptor under anaerobic conditions. Growth experiments of WT and final strains were performed in 3 l BioFlow® 320 reactors (Eppendorf Bioprocess) equipped for the automatic control of pH (with H2SO4 or KOH), temperature, agitation, and nitrogen concentration. Data were directly saved at unit or in BioCommand software. Seed train was prepared 56 h before inoculation in falcon tubes and scaled up to 1 l flasks under anaerobic conditions.

Molecular and genetic techniques

Oligonucleotides used as primers for amplification of DNA fragments were inferred from the working draft genome sequence of M. gryphiswaldense (GenBank accession number CP027526) [80] and purchased from Sigma-Aldrich (Steinheim, Germany). Construction of plasmids was performed by standard recombinant techniques as described in Zwiener et al. [53]. Generated constructs were sequenced by Macrogen Europe (Amsterdam, Netherlands) and sequence data analyzed with Geneious 8.0.5 (Biomatters Ltd).

Construction of markerless gene deletion vectors and mutants

Generation of single and multiple deletion mutants and WT-gusA insertion mutant was carried out by a tailored galK counterselection system as previously reported [53, 81].

We used chromosomal insertion and expression of magnetosome biosynthetic gene clusters, since previous work has shown that episomal expression in M. gryphiswaldense resulted in instability and inhomogenous expression of foreign and magnetosome genes [82, 83]. Multiple deletion mutants were complemented with the pMDJM3 cassette, a recyclable variant of pTpsMAG1 [60] containing lox sites next to the antibiotic marker and all operons necessary for magnetosome formation. For insertion of the chromogenic marker gusA into pMDJM3, RedET recombineering [84] was performed according to BAC Subcloning Kit (Gene Bridges) technical protocols.

Analytical methods

Analysis of putative prophages

Analysis of putative prophages was performed by the phage search tool PHAST [55]. In PHAST, a prophage-like element was considered incomplete if its completeness score was less than 60, questionable if the score was in the range between 60 and 90, and complete if the score was above 90.

Re-sequencing of genomic DNA

Genomic DNA was isolated following the manual instructions of Quick-DNA Midiprep Plus Kit (Zymo Research Europe GmbH). For each isolated gDNA, two sequencing libraries were arranged, one for sequencing on the MiSeq platform (Illumina Inc, NL), and the second for sequencing on the GridION platform [Oxford Nanopore Technologies (ONT), UK]. The former was established using the TruSeq DNA PCR-free Library Kit (Illumina Inc., The Netherlands) and was carried out in a 2 × 300 nt run using a 600 cycle MiSeq Reagent Kit v3 (Illumina Inc, The Netherlands). For ONT sequencing, the Ligation Sequencing Kit SQK-LSK109 was used to arrange the libraries, which were in turn run on a R9.4.1 flow cell. Basecalling of the raw ONT data was carried out with guppy v3.2.8 [85]. For assembly, three assemblers were used: the canu assembler v1.8 [86] was utilized to assemble the ONT data. The resulting assembled contigs were polished applying first the ONT data with racon v1.3.3 [87] and medaka v0.11.5 (Oxford Nanopore Technologies), both relying on minimap2 v2.17-r943 [88] for mapping, followed by switching to the Illumina data and the pilon polisher v1.22 [89] for a total of 10 rounds. For the first 5 rounds, bwa mem [90] was utilized as a mapper, for the final 5 cycles, bowtie2 [91] was applied. In addition, the Illumina data was assembled using newbler v2.8 [92] and both data sets were gathered using unicycler [93]. All assemblies were compared with each other and examined for synteny using r2cat [94]. All three assemblies were combined and manually curated using consed [95]. Annotation of the finished genomes was carried out using prokka v1.11 [96] SNPs and small indels were identified using snippy v4.0 [97] while larger rearrangements were recognized manually using SnapGene (GSL Biotech).

Preparation of samples for transmission electron microscopy (TEM)

For conventional transmission electron microscopy (TEM) of cell and magnetosome morphologies, cultures were grown under microoxic conditions in FSM at 28 °C. Overnight cultures were fixed in 1.5% formaldehyde and adsorbed onto carbon-coated copper-mesh grids (Science Services, Munich, Germany). TEM was performed on a JEOL 1400 (Japan) with an acceleration voltage of 80 kV and micrographs were analyzed using the software ImageJ [98].

Genetic stability assay

To test genetic stability of the reporter gusA, overnight cultures were transferred to 96-well-plates and incubated for ten passages under aerobic conditions alternating between 4 h at 4 °C and 44 h at 28 °C. 12 independent parallels of each strain were plated on FSM agar moistened with 250 µl of a 10 mg/ml X-Gluc stock solution on its surface. Clones producing active GusA could be visually screened by their blue color, while mutations inside gusA resulted in the loss of the ability to cleave X-Gluc, and thus in in white or brownish colonies after 7–10 days of incubation at 28 °C. Colonies were counted at time points 0 (t0) and after ten passages (t10) and mutations identified by PCR and sequencing.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.



Magnetosome Island


Wild type

IS element:

Insertion element


Magnetic nanoparticles


Mitomycin C

tn-tandem :


Kmr :

Kanamycin resistant

Kms :

Kanamycin sensitive


Electron dense particles




dl-α,ε-Diaminopimelic acid


  1. Uebe R, Schüler D. Magnetosome biogenesis in magnetotactic bacteria. Nat Rev Microbiol. 2016;14:621–37.

    Article  CAS  PubMed  Google Scholar 

  2. Lefèvre CT, Bazylinski DA. Ecology, diversity, and evolution of magnetotactic bacteria. Microbiol Mol Biol Rev. 2013;77(3):497–526.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  3. Faivre D, Schüler D. Magnetotactic bacteria and magnetosomes. Chem Rev. 2008;108(11):4875–98.

    Article  CAS  PubMed  Google Scholar 

  4. Sun C, Lee JSH, Zhang M. Magnetic nanoparticles in MR imaging and drug delivery. Adv Drug Deliv Rev. 2008;60(11):1252–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Taukulis R, Widdrat M, Kumari M, Heinke D, Rumpler M, Tompa É, Uebe R, Kraupner A, Cebers A, Schüler D, Pósfai M, Hirt AM, Faivre D. Magnetic iron oxide nanoparticles as MRI contrast agents—a comprehensive physical and theoretical study. Magnetohydrodynamics. 2015;51(4):721–47.

    Article  Google Scholar 

  6. Alphandéry E. Applications of magnetosomes synthesized by magnetotactic bacteria in medicine. Front Bioeng Biotechnol. 2014;2:5.

    PubMed  PubMed Central  Google Scholar 

  7. Vargas G, Cypriano J, Correa T, Leão P, Bazylinski DA, Abreu F. Applications of magnetotactic bacteria, magnetosomes and magnetosome crystals in biotechnology and nanotechnology: mini-review. Molecules. 2018;23(10):2438.

    Article  PubMed Central  CAS  Google Scholar 

  8. Orlando T, Mannucci S, Fantechi E, Conti G, Tambalo S, Busato A, Innocenti C, Ghin L, Bassi R, Arosio P, Orsini F, Sangregorio C, Corti M, Casula MF, Marzola P, Lascialfari A, Sbarbati A. Characterization of magnetic nanoparticles from Magnetospirillum gryphiswaldense as potential theranostics tools. Contrast Media Mol Imaging. 2016;11(2):139–45.

    Article  CAS  PubMed  Google Scholar 

  9. Heinke D, Kraupner A, Eberbeck D, Schmidt D, Radon P, Uebe R, Schüler D, Briel A. MPS and MRI efficacy of magnetosomes from wild-type and mutant bacterial strains. IJMPI. 2017;3(2):1706004.

    Google Scholar 

  10. Kraupner A, Eberbeck D, Heinke D, Uebe R, Schüler D, Biel A. Bacterial magnetosomes—nature’s powerful contribution to MPI tracer research. Nanoscale. 2017;9(18):5788–93.

    Article  CAS  PubMed  Google Scholar 

  11. Lang C, Schüler D. Biogenic nanoparticles: production, characterization, and application of bacterial magnetosomes. J Phys Condens Matter. 2006;18:2815–28.

    Article  CAS  Google Scholar 

  12. Lisy MR, Hartung A, Lang C, Schüler D, Richter W, Reichenbach JR, Kaiser WA, Hilger I. Fluorescent bacterial magnetic nanoparticles as bimodal contrast agents. Invest Radiol. 2007;42(4):235–41.

    Article  CAS  PubMed  Google Scholar 

  13. Schwarz S, Fernandes F, Sanroman L, Hodenius M, Lang C, Himmelreich U, Fama G, Schmitz-Rode T, Schüler D, Hoehn M, Zenke M, Hieronymus T. Synthetic and biogenic magnetite nanoparticles for tracking of stem cells and dendritic cells. J Magn Magn Mater. 2009;321(10):1533–8.

    Article  CAS  Google Scholar 

  14. Hergt R, Dutz S, Röder M. Effects of size distribution on hysteresis losses of magnetic nanoparticles for hyperthermia. J Phys Condens Matter. 2008;20:385214.

    Article  PubMed  CAS  Google Scholar 

  15. Alphandéry E, Guyot F, Chebbi I. Preparation of chains of magnetosomes, isolated from Magnetospirillum magneticum strain AMB-1 magnetotactic bacteria, yielding efficient treatment of tumors using magnetic hyperthermia. Int J Pharm. 2012;434(1–2):444–52.

    Article  PubMed  CAS  Google Scholar 

  16. Mickoleit F, Schüler D. Generation of multifunctional magnetic nanoparticles with amplified catalytic activities by genetic expression of enzyme arrays on bacterial magnetosomes. Adv Biosyst. 2018;2:1700109.

    Article  CAS  Google Scholar 

  17. Mickoleit F, Lanzloth C, Schüler D. A versatile toolkit for controllable and highly selective multifunctionalization of bacterial magnetic nanoparticles. Small. 2020;16(16):e1906922.

    Article  PubMed  CAS  Google Scholar 

  18. Pollithy A, Romer T, Lang C, Müller FD, Helma J, Leonhardt H, Rothbauer U, Schüler D. Magnetosome expression of functional camelid antibody fragments (nanobodies) in Magnetospirillum gryphiswaldense. Appl Environ Microbiol. 2011;77(17):6165–71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Borg S, Rothenstein D, Bill J, Schüler D. Generation of multi-shell magnetic hybrid nanoparticles by encapsulation of genetically engineered and fluorescent bacterial magnetosomes with ZnO and SiO2. Small. 2015;11(33):4209–17.

    Article  CAS  PubMed  Google Scholar 

  20. Mickoleit F, Schüler D. Generation of nanomagnetic biocomposites by genetic engineering of bacterial magnetosomes. Bioinspired Biomim Nanobiomater. 2018;8(1):86–98.

    Article  Google Scholar 

  21. Mickoleit F, Borkner CB, Toro-Nahuelpan M, Herold HM, Maier DS, Plitzko JM, Scheibel T, Schüler D. In vivo coating of bacterial magnetic nanoparticles by magnetosome expression of spider silk-inspired peptides. Biomacromolecules. 2018;19(3):962–72.

    Article  CAS  PubMed  Google Scholar 

  22. Mickoleit F, Jérôme V, Freitag R, Schüler D. Bacterial magnetosomes as novel platform for the presentation of immunostimulatory, membrane-bound ligands in cellular biotechnology. Adv Biosyst. 2020;4(3):e1900231.

    Article  PubMed  Google Scholar 

  23. Zeytuni N, Uebe R, Maes M, Davidov G, Baram M, Raschdorf O, Friedler A, Miller Y, Schüler D, Zarivach R. Bacterial magnetosome biomineralization-a novel platform to study molecular mechanisms of human CDF-related Type-II diabetes. PLoS ONE. 2014;9(5):e97154.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  24. Schleifer KH, Schüler D, Spring S, Weizenegger M, Amann R, Ludwig W, Köhler M. The genus Magnetospirillum gen. nov. description of Magnetospirillum gryphiswaldense sp. nov. and transfer of Aquaspirillum magnetotacticum to Magnetospirillum magnetotacticum comb. nov. Syst Appl Microbiol. 1991;14:379–85.

    Article  Google Scholar 

  25. Schüler D, Monteil CL, Lefèvre CT. Microbe of the month: Magnetospirillum gryphiswaldense. Trends in Microbiol. 2020.

    Article  Google Scholar 

  26. Müller F, Schüler D, Pfeiffer D. A compass to boost navigation: cell biology of bacterial magnetotaxis. J Bacteriol. 2020.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Lohße A, Kolinko I, Raschdorf O, Uebe R, Borg S, Brachmann A, Plitzko JM, Müller R, Zhang Y, Schüler D. Overproduction of magnetosomes by genomic amplification of biosynthesis-related gene clusters in a magnetotactic bacterium. Appl Environ Microbiol. 2016;82(10):3032–41.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  28. Heyen U, Schüler D. Growth and magnetosome formation by microaerophilic Magnetospirillum strains in an oxygen-controlled fermentor. Appl Microbiol Biotechnol. 2003;61:536–44.

    Article  CAS  PubMed  Google Scholar 

  29. Alphandéry E. Bio-synthesized iron oxide nanoparticles for cancer treatment. J Pharm. 2020;586:119472.

    Google Scholar 

  30. Alphandéry E. Applications of magnetotactic bacteria and magnetosome for cancer treatment: a review emphasizing on practical and mechanistic aspects. Drug Discov Today. 2020.

    Article  PubMed  Google Scholar 

  31. Fernández-Castané A, Li H, Thomas ORT, Overton TW. Development of a simple intensified fermentation strategy for growth of Magnetospirillum gryphiswaldense MSR-1: physiological responses to changing environmental conditions. New Biotechnol. 2018;46:22–30.

    Article  CAS  Google Scholar 

  32. Schüler D. Versuche zur Isolierung und Charakterisierung magnetotaktischer Bakterien. Diplomarbeit, Ernst-Moritz-Arndt Universität Greifswald. 1990.

  33. Schüler D, Köhler M. The isolation of a new magnetic spirillum. Zentralblatt Mikrobiologie. 1992;147:150–1.

    Article  Google Scholar 

  34. Baumgart M, Unthan S, Rückert C, Sivalingam J, Grünberger A, Kalinowski J, Bott M, Noack S, Frunzke J. Construction of a prophage-free variant of Corynebacterium glutamicum ATCC 13032 for use as a platform strain for basic research and industrial biotechnology. Appl Environ Microbiol. 2013;79(19):6006–15.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Martínez-García E, Nikel PI, Aparicio T, de Lorenzo V. Pseudomonas 2.0: genetic upgrading of P. putida KT2440 as an enhanced host for heterologous gene expression. Microb Cell Fact. 2014;13:159.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  36. Gödeke J, Paul K, Lassak J, Thormann KM. Phage-induced lysis enhances biofilm formation in Shewanella oneidensis MR-1. ISME J. 2011;5(4):613–26.

    Article  PubMed  CAS  Google Scholar 

  37. Schübbe S, Kube M, Scheffel A, Wawer C, Heyen U, Meyerdierks A, Madkour MH, Mayer F, Reinhardt R, Schüler D. Characterization of a spontaneous nonmagnetic mutant of Magnetospirillum gryphiswaldense reveals a large deletion comprising a putative magnetosome island. J Bacteriol. 2003;185:5779–90.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  38. Ullrich S, Kube M, Schübbe S, Reinhardt R, Schüler D. A hypervariable 130-kilobase genomic region of Magnetospirillum gryphiswaldense comprises a magnetosome island which undergoes frequent rearrangements during stationary growth. J Bacteriol. 2005;187:7176–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Ullrich S, Schüler D. Cre-lox-based method for generation of large deletions within the genomic magnetosome island of Magnetospirillum gryphiswaldense. Appl Environ Microbiol. 2010;76:2439–44.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. McCausland HC, Komeili A. Magnetic genes: studying the genetics of biomineralization in magnetotactic bacteria. PLoS Genet. 2020;16(2):e1008499.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  41. Kolinko I, Jogler C, Katzmann E, Schüler D. Frequent mutations within the genomic magnetosome island of Magnetospirillum gryphiswaldense are mediated by RecA. J Bacteriol. 2011;193:5328–34.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Martínez-García E, Fraile S, Espeso DR, Vecchietti D, Bertoni G, de Lorenzo V. Naked bacterium: emerging properties of a surfome-streamlined Pseudomonas putida strain. ACS Synth Biol. 2020;9(9):2477–92.

    Article  PubMed  CAS  Google Scholar 

  43. Martínez-García E, de Lorenzo V. The quest for the minimal bacterial genome. Curr Opin Biotechnol. 2016;42:216–24.

    Article  PubMed  CAS  Google Scholar 

  44. Fehér T, Birland V, Pósfai G. In the fast lane: large-scale bacterial genome engineering. J Biotechnol. 2012;169(1–2):72–9.

    Article  CAS  Google Scholar 

  45. Baumgart M, Unthan S, Kloß R, Radek A, Polen T, Tenhaef N, Müller MF, Küberl A, Siebert D, Brühl N, Marin K, Hans S, Krämer R, Bott M, Kalinowski J, Wiechert W, Seibold G, Frunzke J, Rückert C, Wendisch V, Noack S. Corynebacterium glutamicum chassis C1*: building and testing a novel platform host for synthetic biology and industrial biotechnology. ACS Synth Biol. 2017;7(1):132–44.

    Article  PubMed  CAS  Google Scholar 

  46. Unthan S, Baumgart M, Radek A, Herbst M, Siebert D, Brühl N, Bartsch A, Bott M, Wiechert W, Marin K, Hans S, Krämer R, Seibold G, Frunzke J, Kalinowski J, Rückert C, Wendisch VF, Noack S. Chassis organism from Corynebacterium glutamicum—a top-down approach to identify and delete irrelevant gene clusters. Biotechnol J. 2015;10:290–301.

    Article  CAS  PubMed  Google Scholar 

  47. Pósfai G, Plunkett G 3rd, Fehér T, Frisch D, Keil GM, Umenhoffer K, Kolisnychenko V, Stahl B, Sharma SS, de Arruda M, Burland V, Harcum SW, Blattner FR. Emerging properties of reduced-genome Escherichia coli. Science. 2006;312(5776):1044–6.

    Article  PubMed  CAS  Google Scholar 

  48. Lieder S, Nikel PI, de Lorenzo V, Takors R. Genome reduction boosts heterologous gene expression in Pseudomonas putida. Microb Cell Fact. 2015;14:23.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Martínez-García E, Jatsenko T, Kivisaar M, de Lorenzo V. Freeing Pseudomonas putida KT2440 of its proviral load strengthens endurance to environmental stresses. Environ Microbiol. 2015;17(1):76–90.

    Article  PubMed  CAS  Google Scholar 

  50. Pfeifer E, Michniewski S, Gätgens C, Münch E, Müller F, Polen T, Millard A, Blombach B, Frunzke J. Generation of a prophage-free variant of the fast-growing bacterium Vibrio natriegens. Appl Environ Microbiol. 2019;85(17):e00853-e919.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Choi JW, Yim SS, Kim MJ, Jeong KJ. Enhanced production of recombinant proteins with Corynebacterium glutamicum by deletion of insertion sequences (IS elements). Microb Cell Fact. 2015;14:207.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  52. Suárez GA, Renda BA, Dasgupta A, Barrick JE. Reduced mutation rate and increased transformability of transposon-free Acinetobacter baylyi ADP1-ISx. Appl Environ Microbiol. 2017;83(17):e01025-e1117.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Zwiener T, Mickoleit F, Dziuba M, Rückert C, Busche T, Kalinowski J, Faivre D, Uebe R, Schüler D. Identification and elimination of irrelevant genomic content for magnetosome biosynthesis by large-scale deletion in Magnetospirillum gryphiswaldense. BMC Microbiol. 2021 (Accepted for publication, 2021).

  54. de Lorenzo V, Krasnogor N, Schmidt M. For the sake of the bioeconomy: define what a synthetic biology chassis is! New Biotechnol. 2021;60:44–51.

    Article  CAS  Google Scholar 

  55. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. PHAST: a fast phage search tool. Nucl Acids Res. 2011;39:W347–52.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Engelhardt T, Sahlberg M, Cypionka H, Engelen B. Induction of prophages from deep-subseafloor bacteria. Environ Microbiol Rep. 2011;3:459–65.

    Article  PubMed  Google Scholar 

  57. Siguier P, Gourbeyre E, Chandler M. Bacterial insertion sequences: their genomic impact and diversity. FEMS Microbiol Rev. 2014;38(5):865–91.

    Article  CAS  PubMed  Google Scholar 

  58. Bazylinski DA, Dean AJ, Schüler D, Phillips EJP, Levley DR. N2-dependent growth and nitrogenase activity in the metal-metabolizing bacteria, Geobacter and Magnetospirillum species. Environ Microbiol. 2000;2(3):266–73.

    Article  CAS  PubMed  Google Scholar 

  59. Araujo ACV, Abreu F, Silva KT, Bazylinski DA, Lins U. Magnetotactic bacteria as potential sources of bioproducts. Mar Drugs. 2015;13(1):389–430.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  60. Dziuba MV, Zwiener T, Uebe R, Schüler D. Single-step transfer of biosynthetic operons endows a non-magnetotactic Magnetospirillum strain from wetland with magnetosome biosynthesis. Environ Microbiol. 2020;22(4):1603–18.

    Article  CAS  PubMed  Google Scholar 

  61. Raschdorf O, Müller FD, Pósfai M, Plitzko JM, Schüler D. The magnetosome proteins MamX, MamZ and MamH are involved in redox control of magnetite biomineralization in Magnetospirillum gryphiswaldense. Mol Microbiol. 2013;89:872–86.

    Article  CAS  PubMed  Google Scholar 

  62. Riese C, Uebe R, Rosenfeldt S, Schenk AS, Jérôme V, Freitag R, Schüler D. An automated oxystat fermentation regime for microoxic cultivation and magnetosome production in Magnetospirillum gryphiswaldense. Microb Cell Fact. 2020;19:206.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Schüler D, Uhl R, Bäuerlein E. A simple light-scattering method to assay magnetism in Magnetospirillum gryphiswaldense. FEMS Microbiol Lett. 1995;132:139–45.

    Article  Google Scholar 

  64. Bartosik D, Sochacka M, Baj J. Identification and characterization of transposable elements of Paracoccus pantotrophus. J Bacteriol. 2003;185(13):3753–63.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Little MS, Ervin SM, Walton WG, Tripathy A, Xu Y, Liu J, Redinbo MR. Active site flexibility revealed in crystal structure of Parabacteroides merdae β-glucoronidase from human gut microbiome. Protein Sci. 2018;27(12):2010–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  66. Hashimoto M, Ichimura T, Mizoguchi H, Tanaka K, Fujimitsu K, Keyamura K, Ote T, Yamakawa T, Yamazaki Y, Mori H, Katayama T, Kato J. Cell size and nucleoid organization of engineered Escherichia coli cells with a reduced genome. Mol Microbiol. 2005;55(1):137–49.

    Article  CAS  PubMed  Google Scholar 

  67. Reuß DR, Commichau FM, Gundlach J, Zhu B, Stülke J. The blueprint of a minimal cell: MiniBacillus. Microbiol Mol Biol. 2016;80(4):955–87.

    Article  Google Scholar 

  68. Ikeda M, Nakagawa S. TheCorynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl Microbiol Biotechnol. 2003;62:99–109.

    Article  CAS  PubMed  Google Scholar 

  69. Suzuki N, Nonaka H, Tsuge Y, Inui M, Yukawa H. New multiple-deletion method for the Corynebacterium glutamicum genome, using a mutant lox sequence. Appl Environ Microbiol. 2005;71(12):8472–80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  70. Tsuge Y, Suzuki N, Inui M, Yukawa H. Random segment deletion based on IS31831 and Cre/loxP excision system in Corynebacterium glutamicum. Appl Microbiol Biotechnol. 2007;74:1333–41.

    Article  CAS  PubMed  Google Scholar 

  71. Leprince A, de Lorenzo V, Völler P, van Passel MWJ, Martins dos Santos VAP. Random and cyclical deletion of large DNA segments in the genome of Pseudomonas putida. Environ Microbiol. 2012;14(6):1444–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  72. Liang P, Zhang Y, Xu B, Zhao Y, Liu X, Gao W, Ma T, Yang C, Wang S, Liu R. Deletion of genomic islands in the Pseudomonas putida KT2440 genome can create an optimal chassis for synthetic biology applications. Microb Cell Fact. 2020;19:70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  73. Zhu D, Fu Y, Liu F, Xu H, Saris PEJ, Qiao M. Enhanced heterologous protein productivity by genome reduction in Lactococcus lactis NZ9000. Microb Cell Fact. 2017;16:1.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  74. Komatsu M, Uchiyama T, Omura S, Cane DE, Ikeda H. Genome-minimized Streptomyces host for the heterologous expression of secondary metabolism. PNAS. 2010;107(6):2646–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  75. Ikeda H, Shin-ya K, Omura S. Genome mining of the Streptomyces avermitilis genome and development of genome-minimized hosts for heterologous expression of biosynthetic gene clusters. J Ind Microbiol Biotechnol. 2014;41(2):233–50.

    Article  CAS  PubMed  Google Scholar 

  76. Bu Q-T, Yu P, Wang J, Li Z-Y, Chen X-A, Mao X-M, Li Y-Q. Rational construction of genome-reduced and high-efficient industrial Streptomyces chassis based on multiple comparative genomic approaches. Microb Cell Fact. 2019;18:16.

    Article  PubMed  PubMed Central  Google Scholar 

  77. Umenhoffer K, Draskovits G, Nyerges Á, Karcagi I, Bogos B, Tímár E, Csörgő B, Herczeg R, Nagy I, Pál C, Pósfai G. Genome-wide abolishment of mobile genetic elements using genome shuffling and CRISPR/Cas-assisted MAGE allows the efficient stabilization of bacterial chassis. ACS Synth Biol. 2017;6(8):1471–83.

    Article  CAS  PubMed  Google Scholar 

  78. Wang X, Kim Y, Ma Q, Hong SH, Pokusaeva K, Sturino JM, Wood TK. Cryptic prophages help bacteria cope with adverse environments. Nat Commun. 2010;1:147.

    Article  PubMed  CAS  Google Scholar 

  79. Sambrook J, Russell DW. Molecular cloning: a laboratory manual. 3rd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2001.

    Google Scholar 

  80. Uebe R, Schüler D, Jogler C, Wiegand S. Reevaluation of the complete genome sequence of Magnetospirillum gryphiswaldense MSR-1 with single-molecule real-time sequencing data. Genome Announc. 2018;6(17):e00309-e318.

    Article  PubMed  PubMed Central  Google Scholar 

  81. Raschdorf O, Plitzko JM, Schüler D, Müller FD. A tailored galK counterselection system for efficient markerless gene deletion and chromosomal tagging in Magnetospirillum gryphiswaldense. Appl Environ Microbiol. 2014;80(14):4323–30.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  82. Borg S, Hofmann J, Pollithy A, Lang C, Schüler D. New vectors for chromosomal integration enable high-level constitutive or inducible magnetosome expression of fusion proteins in Magnetospirillum gryphiswaldense. Appl Environ Microbiol. 2014;80(8):2609–16.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  83. Kolinko I, Lohße A, Borg S, Raschdorf O, Jogler C, Tu Q, Pósfai M, Tompa É, Plitzko JM, Brachmann A, Wanner G, Müller R, Zhang Y, Schüler D. Biosynthesis of magnetic nanostructures in a foreign organism by transfer of bacterial magnetosome gene clusters. Nat Nanotechnol. 2014;9:193–7.

    Article  CAS  PubMed  Google Scholar 

  84. Muyrers JPP, Zhang Y, Stewart AF. Recombinogenic engineering: new options for cloning and manipulating DNA. Trends in Biochem Sci. 2001;23:325–31.

    Article  Google Scholar 

  85. Wick RR, Judd LM, Holt KE. Performance of neural network basecalling tools for Oxford Nanopore sequencing. Genome Biol. 2019;20(1):129.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  86. Koren S, Walenz BP, Berlin K, Miller JR, Bergman NH, Phillippy AM. Canu: scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation. Genome Res. 2017;27(5):722–36.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  87. Vaser R, Sović I, Nagarajan N, Šikić M. Fast and accurate de novo genome assembly from long uncorrected reads. Genome Res. 2017;27(5):737–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  88. Li H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics. 2018;34(18):3094–100.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  89. Walker BJ, Abeel T, Shea T, Priest M, Abouelliel A, Sakthikumar S, Cuomo CA, Zeng Q, Wortman J, Young SK, Earl AM. Pilon: an integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS ONE. 2014;9(11):e112963.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  90. Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. 2013. arXiv:1303.3997.

  91. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9(4):357–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  92. Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bemben L, Berka J, et al. Genome sequencing in microfabricated high-density picolitre reactors. Nature. 2005;437(7057):376–80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  93. Wick RR, Judd LM, Gorrie CL, Holt KE. Unicycler: resolving bacterial genome assemblies from short and long sequencing reads. PLoS Comput Biol. 2017;13(6):e1005595.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  94. Husemann P, Stoye J. r2cat: synteny plots and comparative assembly. Bioinformatics. 2009;26(4):570–1.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  95. Gordon D, Green P. Consed: a graphical editor for next-generation sequencing. Bioinformatics. 2013;29(22):2936–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  96. Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30(14):2068–9.

    Article  CAS  PubMed  Google Scholar 

  97. Seemann T. snippy: fast bacterial variant calling from NGS reads. 2015.

  98. Abràmoff MD, Magalhães PJ, Ram SJ. Image processing with imageJ. Biophotonics Int. 2004;11:36–42.

    Google Scholar 

Download references


This work was supported by European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant No. 692637 to D.S.), and by the Federal Ministry of Education and Research (BMBF) (Grant MagBioFab to D.S. and R. U.). We thank S. Geimer and M. Schüler for assistance with transmission electron microscopy, and J. Mansky for the help in the construction of pMDJM3 plasmid.


Open Access funding enabled and organized by Projekt DEAL. This study was supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant No. 692637 to D.S.), and the Federal Ministry of Education and Research (BMBF) (Grant MagBioFab to D.S. and R.U.).

Author information

Authors and Affiliations



TZ and DS conceived the study and designed the experiments. TZ carried out the genetic manipulations and the characterization of the strains. TZ, DS, FM and RU designed genetic stability assay. MD designed and generated plasmid pMDJM3. CR, TB and JK performed resequencing of strains. TZ and DS wrote the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Dirk Schüler.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors have read and approved the current version of the manuscript.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1 of “Towards a ‘chassis’ for bacterial magnetosome biosynthesis: genome streamlining of

Magnetospirillum gryphiswaldense by multiple deletions”: Table S1. Overview of all single deletion mutants which were also combined in strains ∆TZ-16 and ∆TZ-17. Table S2. Overview of primers used in this study. UF = upstream forward; UR = upstream reverse; DF = downstream forward; DR = downstream reverse. Figure S1. Molecular organization of nif operon in M. gryphiswaldense. The deleted nitrogen fixation cluster comprises 16 genes necessary for nitrogen fixation (shown in red): nifWABZTHDK, fixABC, draGT and three ferredoxins (MSR1_18560; MSR1_18600; MSR1_18640). Black arrows represent other genes encoding a putative rubrerythrin protein (MSR1_18580), a SIR2-like domain containing protein (MSR1_18630), a GAF domain-containing protein (MSR1_18650), a biliverdin-producing heme oxygenase (MSR1_18660) and a tRNA (MSR1_18670). Figure S2. Phenotypic characterization of multiple deletion mutants. Electron micrographs of combinatorial deletion mutants ∆TZ-01–∆TZ-15. Scale bars: left columns 500 nm; right columns 100 nm. Cell growth of strains ∆TZ-01–∆TZ-15 under aerobic and anaerobic conditions as well as oxidative stress (H2O2) and moderate heat stress (33 °C). Each strain was analyzed in triplicates and each curve shows the average. Figure S3. Genetic organization of the Kmr, false positive mutant ∆TZ-15∆M04 K752 Kmr. The targeted M04 had not been deleted but was still maintained in the genome. A large part (~ 9.1 kb) of the 10.2 kb deletion vector pORFM-GalK-M04 harboring the Kmr gene was found to be inserted at the intended site, but harboring a spontaneous duplication of both the upstream and downstream homologous regions intended for targeted insertion of the deletion construct by homologous recombination. In addition, the galK gene was inactivated by insertion of a copy of the IS element ISMgr2 into the central region. Figure S4. Growth profiles of these strains induced with MMC with concentrations between 0.2 and 0.3 µg/ml MMC, induced 8 h. Cells were washed twice, adjusted to initial OD and growth experiments started at 28 °C under aerobic conditions and each strain was analyzed in triplicates while each curve shows its average (standard deviation < 5%). Figure S5. Experimental procedure of the genetic stability assay (A) and identified magnetosome phenotypes (B). Blue arrows indicate blue colonies while white/brown colonies are marked by black arrows (A). TEM micrographs (B) show WT-like magnetosome chains (upper micrograph) and flake-like particles (lower micrograph, white arrows) could be observed in several of WT-gusA clones after ten passages. Scale bars: 100 nm.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zwiener, T., Dziuba, M., Mickoleit, F. et al. Towards a 'chassis' for bacterial magnetosome biosynthesis: genome streamlining of Magnetospirillum gryphiswaldense by multiple deletions. Microb Cell Fact 20, 35 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: