On the backward excursions in the free-swimming magnetotactic multicellular prokaryote ‘Candidatus Magnetoglobus multicellularis’

Abstract

Magnetotactic bacteria align to magnetic field lines while swimming in a behavior known as magnetotaxis. They are diverse phylogenetically and morphologically and include both unicellular and multicellular morphologies. The magnetotactic multicellular prokaryote (MMP) ‘Candidatus Magnetoglobus multicellularis’ has been extensively studied, even though it remains uncultured up to now. It swims back and forth along magnetic field lines, exhibiting a preferential swimming direction that is usually south-seeking, as described for most magnetotactic microorganisms from the Southern Hemisphere. In order to understand the effects of the magnetic field intensity on the backward excursions of ‘Ca. M. multicellularis’, we applied magnetic fields ranging from 0.09 to 3.4 mT and recorded their movements. Each microorganism was followed frame by frame generating position coordinates, which were used to calculate the frequency of reversal events, as well as the time, distance, and velocity. The velocities in forward movements before and after backward excursions are similar, but no relation was found with the velocity in backward movements. The shapes of the trajectories are distinct in forward and backward movements. In addition, the backward velocities are usually higher. The sharp changes in direction (approximately 180°) indicate that reversal of the flagella rotation direction is the probable mechanism for swimming backward. In conclusion, the backward excursions provide additional freedom of movement to the microorganism, especially when it is constrained by magnetic fields stronger than the Earth’s. Backward movements integrate the ‘Ca. M. multicellularis’ behavioral toolbox, which includes also negative phototaxis, photokinesis, magnetotaxis and possibly helical klinotaxis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-024-01584-8.

Introduction

Prokaryotes swim in aquatic environments using helical, rotatory flagella. Flagella propel the cell body forward and is also responsible for changes in the swimming direction, producing at least seven distinct trajectory patterns, which include run-and-tumble, run-and-reverse, and helical klinotaxis. Swimming in prokaryotes is usually a result of complex intracellular signaling that integrates chemical (e.g., nutrients) and physical (e.g., light) signals into a single behavioral response [1]. In some bacteria, this behavioral response also integrates passive alignment to magnetic field lines in a behavior known as magnetotaxis [2–6].

Magnetotaxis is caused by intracytoplasmic, membrane-bound magnetic nanocrystals known as magnetosomes [6–8]. Magnetosomes are attached to cytoskeletal elements and impart whole microorganisms with a magnetic moment, which interacts with the local magnetic field, leading to the alignment of the swimming trajectory to the magnetic field lines [6]. In the environment, magnetotaxis can be used to navigate up and down due to the vertical component of the locally inclined Earth’s magnetic field [2–4, 6, 9–11].

Magnetotaxis is coupled to photo-, chemo- and aerotaxis, which results in complex behaviors, most of which involve variations in the run-and-reverse pattern [2–6, 10, 12–29]. For example, Magnetospirillum magnetotacticum MS-1 uses a run-and-reverse pattern to form and maintain aerotactic bands [2, 3, 9], whereas Magnetospirillum magneticum AMB-1 has been shown to use a series of runs and reversals to escape from flat surfaces [25]. Magnetococcus marinus MC-1 is microaerobic and swims back and forth along magnetic field lines according to the concentrations of dissolved O₂. Above a lower threshold of O₂ concentration, M. marinus swims downward, and below a higher threshold, it swims upward. Thus, it tends to swim between two thresholds within the range corresponding to the best concentrations for survival and growth [2, 3].

Magnetotaxis and magnetosomes are the behavioral and structural hallmarks of a phylogenetically diverse group of bacteria known as magnetotactic bacteria [6, 11, 14, 19, 23, 24, 26, 29–32]. The morphologies of magnetotactic bacteria include cocci, rods, vibria, spirilla and multicellular forms collectively known as multicellular magnetotactic prokaryotes (MMPs) [4, 6, 10–12, 17, 19, 23, 24, 26, 27, 29, 31–39]. MMPs are monophyletic, clustering within a large group within the δ-proteobacteria [14, 17, 19, 23, 24, 26, 29, 31, 37, 39]. Spherical and ellipsoidal forms have been observed, all of which share the basic body plan, which is a monolayer of cells around a small void [26, 27, 29, 34, 35, 39–41]. Flagella are abundant and distributed on the outer surface [11, 19, 23, 24, 40–42]. All known MMPs proliferate by binary fission, i.e., each multicellular individual divides into two equivalent offspring [19, 24, 26, 27, 36, 41]. Genomic analyses of spherical MMPs suggest that they are anaerobes or microaerophiles that use H₂ and simple organic acids as carbon sources/electron donors and sulfate and/or O₂ as the final electron acceptors [39]. No MMP has been axenically cultured until now; thus, all studies have been performed with samples obtained from environmental samples, microcosms, or enrichment cultures [4, 6, 10–12, 14, 17, 19, 23, 24, 26, 27, 29, 31–42].

The best studied MMP to date is ‘Candidatus Magnetoglobus multicellularis’, which is spherical [5, 14, 16, 20–22, 28, 35, 36, 41–45]. In ‘Ca. M. multicellularis’, the flagellar filaments are short relative to the body of the microorganism, which precludes the formation of a single flagellar bundle during swimming [42]. They swim on average at 52 to 112 μm/s [5, 20, 21, 28, 42, 44]. ‘Ca. M. multicellularis’ swim in helical trajectories, where both the body and the trajectory rotate clockwise [14–16]. At low intensity magnetic fields, ‘Ca. M. multicellularis’ swims in trajectories with rather sinuous axes. Increasing the magnetic field intensity improves the alignment and results in straighter paths [5]. Because the magnetic moment provided by magnetosomes is sufficient for good alignment at low-intensity magnetic fields, such as Earth’s [43], it was suggested that poor alignment of ‘Ca. M. multicellularis’ could be due to chemo- and/or phototaxis [5].

The predominant swimming direction of magnetotactic microorganisms under oxygenated conditions is parallel to magnetic field lines (north-seeking) in the Northern Hemisphere and antiparallel (south-seeking) in the Southern Hemisphere [3–6, 8, 11, 14, 15, 17, 19–21, 23, 24, 26, 27, 31, 35, 37, 43, 46], with a few exceptions (e.g., [9]). In addition to forward movement, MMPs exhibit intermittent backward excursions, which are called “backward excursions”, “ping-pong” or “escape motility”. Backward excursions have been observed during free swimming or when MMPs are bound by an obstacle, such as the border of a water drop or a glass coverslip [10–17, 19, 23, 24, 26–29, 31, 37, 46, 47]. In ellipsoidal MMPs, the motility axis is parallel to the long axis of the body. Accordingly, precession occurs around their long axes. For the backward excursions, they reverse the direction of movement without reversing the microorganism body, which maintains alignment relative to the magnetic field [19, 24, 26, 27]. Observation of moving ‘Ca. M. multicellularis’ shows that they behave the same way [16]. Under high-intensity blue, violet, and/or UV light, whole populations swim backward, indicating the role of backward excursions in escaping dangerous situations [11, 18–20, 23, 24, 26, 27]. It was speculated that this light effect could be a response to increased radical oxygen species (ROS) [27]. Accordingly, backward excursions were not observed under anoxia [46].

Most studies on the backward excursions of MMPs have been done with microorganisms concentrated at the air-water interface [13, 18, 28, 46], a condition uncommon in their natural environment. When concentrated at borders, movements are physically constrained, but at low-intensity magnetic fields ( ≤ ~ 1 mT), the MMP ‘Ca. M. multicellularis’ moves laterally in seemingly aleatory directions [16]. Under such conditions, the identification of short backward excursions is hampered, and the effects of magnetic fields on backward movements can become confusing. In addition, there are conflicting views on the effects of the magnetic field intensity on backward excursions [13, 28].

In this work, we analyzed several aspects of the backward excursions in ‘Ca. M. multicellularis’ that occurs during free motion under applied magnetic fields ranging from 0.09 to 3.4 mT. Within this range, the alignment of ‘Ca. M. multicellularis’ to the magnetic field lines increases with the magnetic field intensity [5, 16]. We assessed the changes in velocity, as well as the distances traveled and time. Our aim was to eliminate the effects of borders and unravel the effects of magnetic fields on these movements.

Materials and methods

Water and sediment samples were collected at Araruama Lagoon, Rio de Janeiro State, Brazil (22° 55’ 24” S, 42°18’12” W). They were maintained in the laboratory in 1 L plastic bottles at room temperature under dim light for a few weeks. Periodically, about 100 mL of surface sediments and 100 mL water were retrieved from a single bottle and subjected to applied magnetic fields in order to retrieve magnetotactic microorganisms as described previously [5, 16, 48].

To obtain light micrographs, a drop of lagoon water containing ‘Candidatus Magnetoglobus multicellularis’ was placed on a glass slide and covered by a coverslip. The samples were subsequently observed and imaged with a Zeiss Axioplan 2 microscope adjusted to Nomarski interference contrast.

To record the movements, samples of lagoon water enriched in ‘Ca. M. multicellularis’ (> 100 individuals/100 µL) were placed on a glass slide between two pieces of adhesive tape to create a gap of approximately 60 µm between the slide and the coverslip. This gap is important for minimizing the effects of glass surfaces on swimming microorganisms. The water containing the microorganisms did not contact the tape.

A light microscope (Bioval L2000C) was used to record the movements of the magnetotactic microorganisms under applied magnetic fields. An objective lens with 20x magnification was selected for imaging. A microscope stage was built with plastic materials and adapted to the microscope to hold coiled coils approximately in the Helmholtz configuration and aligned parallel to the plane of focus of the microscope as described previously [5]. The magnetic field intensities provided by the coils were 0.09, 0.15, 0.28, 0.5, 1.0, 2.0, 2.9 and 3.4 mT, as measured by two different Gaussmeters (Global Mag TLMP-HALL-050 or Global Mag TLMPHALL-05 K-T1, depending on the magnetic field intensity range). Within this range, the alignment of ‘Ca. M. multicellularis’ increases with magnetic field intensity and approaches saturation [5]. Video recording and processing were performed as described previously [5, 16] using ImageJ software [49], generating worksheets that were further processed in Microsoft Excel software.

The backward excursions were recognized as sudden reversals in the swimming direction approaching 180°. Only backward excursions of 0.13 s (4 frames) and longer could be recognized as such. Thus, this was the lower limit used throughout this work. Shorter events could not be identified due to noise. The higher limit was set by the record frame, which was 334 × 445 μm. The number of reversals was evaluated by manually counting the number of reversals and microorganisms in three randomly chosen series of video records, each obtained from a single sample. The results are expressed as the number of reversals divided by the number of microorganisms in each video divided by the time in minutes.

We evaluated the distance traveled and the time spent in backward excursions and used them to calculate the speed before, during and after spontaneous reversals of movement direction. For time calculations, each frame was considered to be 1/30 s. For the evaluation of velocity after reversal, data obtained 1 s or more after returning to forward movement were used, after the acceleration approached 0. Since the trajectories are very elongated (radius/pitch ≈ 0.04) [20], the lengths of most sinusoidal projections of the trajectories are very close to the length of the helical paths and to the length of the trajectory axis. In addition, several paths were too short to fit a sinusoidal curve and extract the trajectory parameters. Thus, the distance traveled by the microorganisms before and after reversal of movement direction was evaluated without fitting the sinusoidal curves using the following equation:

1

where x and y are the coordinates for the position in each frame and n refers to the frame number. The velocity was evaluated as follows:

2

where Δt is the time period.

Results and discussion

‘Candidatus Magnetoglobus multicellularis’ showed the usual morphology, as illustrated in Fig. 1, and swam actively along magnetic field lines. Intermittently, they reversed the direction of movement (online resource 1), as reported previously, in a behavior called backward excursions, escape motility, or ping-pong [10–17, 19, 23, 24, 26–29, 31, 37, 46, 47].

Fig. 1.

Nomarski differential interference contrast light microscopy of ‘Ca. M. multicellularis’. Scale bar = 10 mm

The counts of backward excursion events showed highly variable rates, ranging from 0 to 2 events.min^− 1 for each microorganism, with a median of 0.27 events.min^− 1 (n = 3 samples). These values are very close to those found for ‘Ca. M. multicellularis’ and other MMPs when packed close to the border of a water drop [13, 28]. No consistent relationship was observed between the magnetic field intensity and reversal rates (Fig. 2), which agrees with the work of Sepulchro et al. [28].

Fig. 2.

Frequencies of backward excursion events under different magnetic fields in three samples. The results are expressed as the average number of reversals per minute performed by the microorganisms

Figure 3 shows the trajectories of several microorganisms before, during, and after the backward excursions. The shape of each trajectory differs between forward and backward movements: whereas forward movements are clearly helical, the tracks of the backward excursions frequently seem straight. Considering that helical movements are almost universal among microorganisms [50], these apparently straight trajectories may result from large pitches and/or small trajectory radii. Indeed, Sepulchro et al. [28] measured the trajectory radii and frequencies in both backward and the following forward trajectory and found smaller average radii in the backward tracks. Because we analyzed only backward excursion events occurring during free motion, we identified several short events (see Fig. 3e-f and the online resource 1, 1 mT), most of which were too short to fit a sinusoidal curve. Since the trajectories of ‘Ca. M. multicellularis’ are elongated [5, 20], we decided to quantify the trajectory lengths without fitting them to helical curves.

Fig. 3.

Trajectories of ‘Ca. M. multicellularis’ before, during and after backward excursions. “B” indicates the direction of the applied magnetic field. The time between adjacent dots is 1/30 s. Forward movements (light blue) occur from left to right; then, the microorganisms move backward for a while (red); and afterwards, they resume forward movement (dark blue). Some trajectories show two reversal events (e.g., trajectory at the top in “c”)

The alignment of the forward trajectories to the magnetic field lines increased with magnetic field intensity (Fig. 3, blue tracks), as described previously [5, 28, 45]. On the other hand, the initial stretches of both the backward excursions and the subsequent forward movements often showed poor alignment to the magnetic field lines, but if they were long enough, the microorganisms turned and eventually became aligned with the magnetic field lines. Thus, backward excursions change the direction of movement by angles of approximately 180° twice, but because each change in direction is limited by the intensity of the magnetic field, the speed, and the time swimming backwards, the effect on the shape of the trajectory varies widely. The time and distance traveled before realignment to the magnetic field lines depend on the interaction between the magnetic moment of the microorganism and the local magnetic field, the microorganism diameter, and the temperature and viscosity of the medium [51]. From the perspective of ecophysiology, short backward excursions could be used to change the direction of movement (see Fig. 3a, c-f), whereas longer excursions could be useful for escaping from concave grains and magnetic particles (see [16]).

Figure 4 shows plots of ‘Ca. M. multicellularis’ trajectories and their corresponding velocities over time. The velocity before backward excursions is approximately continuous over time (Fig. 4b, d, f, h, j), as in free motion [5, 20]. In the velocity x time plots (Fig. 4b, d, f, h, j), the backward excursions are observed as velocity peaks, preceded by a brief stop or a sharp decrease in velocity for 0.03–0.27 s (1–9 frames). Close examination revealed that most backward excursions begin with sharp acceleration until they reach a peak velocity usually higher than the forward velocity. After some time at the peak velocity, the microorganisms decelerate until the backward excursion ends in a second, longer stop or short period of slow movement (0.03–1.63 s). Forward movement resumes with acceleration for 0.13–0.9 s, when the microorganism reaches a forward velocity similar to that observed before the backward excursion. Similar patterns of deceleration-stop-acceleration to resume forward movement have been shown for both spherical and ellipsoidal MMPs [13, 24, 26, 28, 46]. Qian et al. [26] showed two velocity x frame plots very similar to those of velocity x time in Fig. 4 (this work), indicating that the general pattern for backward excursions is largely the same in ‘Ca. M. multicellularis’ (this work) and ellipsoidal MMPs [26].

Fig. 4.

Plots of trajectory shapes (a, c, e, g, i) and their corresponding velocities over time (b, d, f, h, j) before (light blue), during (red), and after (dark blue) backward excursions. The applied magnetic field intensities were 0.28 mT (a–b), 0.50 mT (c-d), 1.0 mT (e-f), 2.0 mT (g-h) and 2.9 mT (i-j). The time between adjacent points in (a), (c), (e), (g) and (i) is 1/30 s. “B” indicates the direction of the applied magnetic field

Some backward excursions surpassed the frame length (445 μm), precluding measurements of some parameters of those trajectories. Nevertheless, most of them were much smaller (Fig. 5a), with a median of 55 μm for trajectory length. The distribution of time spent swimming backward was similar to that of length (Fig. 5b), peaking at 0.43 s (median). Figure 6 shows a plot of time versus distance traveled in backward excursions. The narrow distribution of the data points and Pearson’s coefficient (r = 0.90) indicate a very high positive correlation. Neither the distance traveled nor the time seemed to depend on the magnetic field intensity (Fig. 6). The onset distance observed in previous work [13, 28] was not observed in the present work, probably due to the observation of backward excursions occurring while microorganisms move freely, which enabled the recognition of very short events as backward excursions.

Fig. 5.

Histograms of the distributions of (a) travel distance and (b) time during backward excursions

Fig. 6.

Correlations between time and distance traveled in backward excursions

Figure 7 shows histograms of the mean velocity before, during and after backward excursions, as well as the peak velocity during backward excursions. ‘Ca. M. multicellularis’ swam at 32 to 221 μm/s (median 118 μm/s, N = 92) before the backward excursions (Fig. 7a) and at 45–201 μm/s (median 135 μm/s, N = 70) after it returns to forward movements (Fig. 7c; Table 1). These values agree with those shown by Sepulchro et al. [28] for forward movements after reversal. In addition, they are very similar to each other and comply with previous work on forward movements of ‘Ca. M. multicellularis’ [5, 20–22, 44, 45]. The mean velocity in forward movements before and after backward excursions seems to be weakly dependent on the magnetic field intensity (Fig. 7a, c), as described previously [20, 22, 45]. On the other hand, no influence of the magnetic field was observed on either the mean velocity or peak velocity during the backward excursions (Fig. 7b, d).

Fig. 7.

Histograms of the distributions of the mean velocity before (a), during (b), after (c), and during the peak velocity (d) in backward excursions under applied magnetic fields ranging from 0.15 to 3.4 mT

Table 1.

Forward velocity (v_f), peak backward velocity (v_b), velocity ratio v_f/v_b, and the ratio of mechanical power (v_f/v_b)². The minimum (min), average (avg) and maximum (max) values are shown

Applied magnetic field (mT)	Forward velocity (vf) (µm/s)			Peak backward velocity (vb) (µm/s)			Velocity ratio vb/vf			Mechanical power ratio (vb/(vf)2			N
	Min	Avg	Max	Min	Avg	Max	Min	Avg	Max	Min	Avg	Max
0.15	36	104	220	185	257	406	1.12	3.50	8.23	1.25	17.8	67.8	6
0.28	32	82	128	96	200	412	1.41	2.46	3.55	1.99	6.67	12.63	10
0.50	65	121	200	38	179	392	0.26	1.96	3.33	0.065	4.77	11.1	10
1.0	57	121	162	44	192	399	0.28	1.89	4.77	0.079	5.23	22.8	13
2.0	40	116	203	49	162	308	0.34	1.41	2.99	0.12	2.51	8.95	19
2.9	89	158	221	34	187	527	0.37	1.24	3.76	0.14	2.06	14.2	24
3.4	61	101	186	76	173	308	0.88	1.82	2.99	0.78	3.88	8.94	5

The peak velocity in the backward excursions varies widely (34–527 μm.s^− 1, median 185 μm.s^− 1, N = 96) (Fig. 7d; Table 1). Usually, this velocity is higher than the velocities before and after the event, as described previously [23, 24, 28, 37]. Indeed, the ratio between the peak velocity during reversal and the forward velocity before reversal varied from 0.26 to 8.23 (median = 1.47, N = 87). The mean velocity during backward excursions (32–275 μm.s^− 1, median 119 μm.s^− 1, N = 90) is similar to the forward velocity measured before and after (Fig. 7a-c), which is probably a coincidence caused by the stops at the beginning and the end (Fig. 4), combined with the short times and distances traveled in most backward excursions (Fig. 5).

The plot of forward velocity before and after backward excursions showed a narrow distribution (Fig. 8a). Accordingly, the Pearson correlation coefficient of r = 0.91 indicates a very high positive correlation. Both confirm the full recovery of forward movements after 0.03–1.63 s of acceleration (see Fig. 4). In contrast, negligible or no correlation was observed between velocity before and during the backward excursions (r = 0.16) or between velocity in forward movement before the event and peak velocity during backward excursions (r = 0.05) (Fig. 6b-c). The very high correlation between the velocity before and after the excursions and the lack of correlation between the velocity at excursions and the velocity before (Fig. 8b-c) and after (not shown) the excursions indicate that different mechanisms are working to move ‘Ca. M. multicellularis’ forward and backward. The different shapes observed in forward and backward trajectories (Figs. 3 and 4), as well as the difference in trajectory radii measured by Sepulchro et al. [28], point in the same direction. As flagella seem to be distributed homogeneously on the surface of the ‘Ca. M. multicellularis’ [42], these different mechanisms may essentially involve rotation of the same flagella in opposing directions. Indeed, Yang et al. [46] reported an abrupt and highly coordinated change in flagellar movements during reversals of movement direction in spherical MMPs. The bacterial flagella exhibit distinct properties when rotating clockwise and counterclockwise, including changes in filament shape, torque, and frequency [52, 53]. Such differences could account for the differences in forward and backward movements reported here.

Fig. 8.

Correlations of mean velocity before and after backward excursions (a), before and during backward excursions (b), and between mean velocity before excursions and peak velocity during excursions (c) under applied magnetic fields ranging from 0.15 to 3.4 mT

In the low-Reynolds number regime, as is the case for swimming bacteria, the inertia of the bacterium can be disregarded. Additionally, considering that the frictional force due to movement is proportional to the velocity and that this force equals the propulsion force of the flagella in the modulus, we can consider that the flagellar output power P is proportional to the square of the bacterial velocity.

3

where P is the flagellar output power; E is the energy; and F is the frictional force applied to the aqueous medium, which equals the propulsion force of the flagellum in the modulus.

Table 1 shows the range of forward and backward velocities along with the velocity ratios and flagellar output power ratios, which were calculated according to Eq. 3. The flagellar output power ratio varied widely (from 0.12 to 67.8, median 2.17) and apparently did not depend on the magnetic field intensity. This very high value (67.8) was obtained from a microorganism swimming forward at 36 μm.s^− 1 and backward at 296 μm.s^− 1. For the trajectory in Fig. 4h, the ratio of the mechanical power associated with the backward excursion is approximately 7.9 times that associated with forward movement (the backward velocity is 2.8 times the forward velocity). Since the flagellar output power reflects the amount of energy spent swimming, these numbers illustrate the importance of backward swimming for ‘Ca. M. multicellularis’. Keim et al. [16] suggested that backward movements could be used to escape magnetic particles in sediments. Yang et al. [46] noticed that backward excursions were observed only if O₂ was allowed to diffuse into water, indicating that this behavior was used to avoid high O₂ concentrations. In addition, it could be used to check if the direction of movement imposed by the applied magnetic field is in agreement with the ecophysiological needs for bacterial survival. By moving faster than usual, bacteria can evaluate differences in environmental conditions along their path, which contributes to survival in the chemical gradients where they thrive.

If there were no mechanisms to swim in directions different from those coinciding with the magnetic field lines, magnetotactic microorganisms could easily be trapped within concave surfaces or around natural magnetic grains in the sediments [16]. Qian et al. [26] proposed that backward excursions in ellipsoidal MMPs would play a role similar to that of tumbles in the run-and-tumble movements of E. coli, which would provide stochastic changes in direction. In the case of magnetotactic microorganisms, any stochastic movements are limited by the local magnetic field [5]. Thus, a kind of run-and-reverse pattern provides additional freedom of movement to magnetotactic microorganisms, independently of magnetic field intensity.

Previous work interpreted some data on the backward excursions of MMPs as evidence for magnetoreception [13, 28], whereas Yang et al. [46] considered classic magnetotaxis sufficient to explain their data. Magnetoreception must involve both sensing and processing of magnetic information. Magnetosomes are considered magneto-sensors, since they convey information of both direction and intensity of magnetic fields to the microorganism’s body [3, 6, 8, 55, 56]. The processing of sensory information can be very simple, for example as described for chemotaxis in E. coli [54]. In ‘Ca. M. multicellularis’, the characteristics of the backward excursions can be explained purely based on the physics of the interaction of the magnetosomes with the applied magnetic field while the microorganism swims, in much the same way as proposed by Nogueira and Lins de Barros [55] and Yang et al. [56] for unicellular magnetotactic bacteria. The lack of effects of magnetic field intensity on the frequency, length, time and speed of backward excursions suggest that they are controlled by something else. Accordingly, Yang et al. [46] proposed that the backward excursions would be a response to oxygen concentrations.

Conclusions

The backward excursions are characterized by a brief stop followed by acceleration until a peak velocity is reached, followed by deceleration until the backward excursion ends in a second stop. Then the microorganisms resume swimming in the forward direction and accelerates until they reach its usual forward velocity. The double reversal by 180° introduces changes in the swimming pathway, which are limited by the intensity of the magnetic field, the speed of the microorganism, and the time spent in the movement – much of the same parameters that affect alignment to the magnetic field lines in the forward movement. In most cases, the flagellar output power is much higher in backward excursions than in forward movements, reflecting the importance of backward swimming for ‘Ca. M. multicellularis’. The frequency, time, distance traveled, and velocity during backward excursions are not affected by the intensity of the magnetic field, at least in the range used in this work. This independence makes them useful for microorganisms to escape high O₂ concentrations [46], high intensity UV light [18–20, 23, 24, 27] and other deleterious conditions in magnetic and/or concave traps under a range of magnetic field intensities [16]. Overall, the movements of ‘Ca. M. multicellularis’ integrate back and forward movements, along with subtle changes in swimming direction, all of which are modulated by the interaction of the magnetic moment of the microorganism with the external magnetic field.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

All the authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Carolina Neumann Keim. The first draft of the manuscript was written by Carolina Neumann Keim, and Marcos Farina commented on previous versions of the manuscript. All the authors read and approved the final manuscript.”

Declarations

Competing interests

The authors have no competing interests to declare that are relevant to the content of this article.