The motility of bacteria is an important factor in their infectivity. In this study, the motility of Leptospira, a member of the spirochete family that causes a zoonotic disease known as leptospirosis, was analyzed in different viscous or osmotic conditions. Motility assays revealed that both pathogenic and saprophytic strains increase their swimming speeds with increasing viscosity. However, only pathogenic Leptospira interrogans maintained vigorous motility near physiological osmotic conditions. This suggests that active motility in physiological conditions is advantageous when Leptospira enters hosts and when it migrates toward target tissues.
Leptospira are spirochetes and include both pathogenic and saprophytic species. The pathogenic species causes leptospirosis, which is a global zoonosis. Leptospira cells colonize the kidneys of rodents, which are the main carriers, and are shed into the environment in urine. Susceptible hosts such as humans are infected via contaminated water . However, the transmission mechanism has not yet been fully elucidated.
Various environmental conditions that could affect their physiology and infective efficiency change drastically during the life cycle of Leptospira. Whereas Leptospira species can live for quite a long time in water , their viability or physiological activity is affected by temperature, pH, osmolarity and viscosity. High temperatures  or acidic conditions [2, 3] markedly reduce the survival probability of Leptospira. Recently, it was reported that physiological osmolarity enhances the degree of expression of two bacterial immunoglobulin-like proteins, LigA and LigB, and that these proteins facilitate the attachment of leptospiral cells to host tissues [4, 5].
Addition of viscous reagents is known to activate bacterial motility to a certain extent . Also, non-motile mutants of some bacterial species have attenuated infectivity [7, 8]. These findings suggest that motility is an important factor in bacterial infectivity. Leptospira species swim in liquids by rotating their coiled cell bodies . Motility assays of Leptospira biflexa have shown that saprophytic Leptospira cells exhibit positive responses to a gradient of viscosity , the swimming speed monotonically increasing with increasing viscosity . Therefore, the motility of Leptospira is thought to play a critical role in invasion of animals or migration within them. However, L. biflexa cells are reportedly immobilized by hypertonic solutions, for example, 0.5 and 1.0 M sucrose solutions . As mentioned above, pathogenic Leptospira cells are excreted in the urine . Though the osmolarity of urine differs among species of animals and also depends on the health of the animals, several reports have shown that the urinary osmolarity of some rodent species is about 1000 mOsm/kg [13, 14], which is approximately equivalent to the osmolarity of a 0.5 M NaCl solution. These findings suggest that the motility of Leptospira might not be responsible for infection after the cells have been excreted in the urine. In this study, the effects of changes in viscosity and osmolarity on the motility of pathogenic and saprophytic Leptospira species were investigated.
In this study, a pathogenic species, Leptospira interrogans serovar manilae strain UP-MMC-NIID , and a saprophytic species, L. biflexa strain Patoc I, were used. Cells were grown in Ellinghausen–McCullough–Johnson–Harris liquid medium at 30°C to stationary phase. The cells were harvested by centrifugation at 8000 g for 2 min at room temperature and resuspended in motility medium. The motility medium was 20 mM potassium phosphate buffer (pH 7.6). NaCl and MC were added to increase the osmolarity and viscosity, respectively. Motility was observed using a dark-field microscope (BH-2; Olympus, Tokyo, Japan) at 23°C and images recorded with a charge coupled device camera (SSC-M350; Sony, Tokyo, Japan) and a DVD recorder (DVR-525H; Pioneer, Kanagawa, Japan).
The motile forms of Leptospira are roughly classified into translation and rotation types . Translational cells exhibit smooth and directional forward or backward swimming by rotating their cell bodies, whereas rotating cells only rotate in one position without translational motion. For analysis of the motile fraction in this study, both translational and rotating cells were counted as motile cells. For measurement of the swimming speed, appropriate sections of the recordings were captured on a computer and the swimming trajectories of individual translating cells were determined by using ImageJ (National Institute of Health, Bethesda, MD, USA). The swimming speeds were calculated from the trajectories by using Microsoft Excel.
The swimming speeds of saprophytic L. biflexa increased in the presence of MC (P < 0.001, Student's t-test; Fig. 1), which is in agreement with a previous report . Pathogenic L. interrogans swam faster in highly viscous liquid (P < 0.001, Student's t-test), which supports the hypothesis that motility facilitates establishment of leptospiral infections. The average speeds of L. interrogans were slightly different from those of L. biflexa in the presence and absence of MC. As predicted by a previous mathematical model , differences in morphological variables or characteristics of the flagellar motor may affect swimming speeds.
Effects of osmolarity on the motility of L. interrogans and L. biflexa near physiological conditions were analyzed in the presence of 150 or 300 mM NaCl. The results are shown in Figure 2. L. interrogans maintained almost the same degree of motility in the presence of 150 mM NaCl as in the absence of NaCl (Fig. 2a). In the presence of 150 mM NaCl, the swimming speed of L. interrogans slightly decreased as a result of a reduction in the number of translating cells, however, vigorous rotation was retained (data not shown). In contrast, the motility of L. biflexa was severely impaired by the addition of 150 mM NaCl, though it was maintained for 24 hr in motility medium without any nutrients or NaCl (Fig. 2b). Motility attenuated in media containing NaCl was not restored immediately by removal of NaCl (data not shown). Thus, pathogenic Leptospira are resistant to osmotic change, whereas saprophytic ones are sensitive to it. These results indicate that resistance to osmotic change may be involved in the infection mechanisms of Leptospira.
In this study, the motility of pathogenic and saprophytic Leptospira species in different viscosity and osmotic conditions were analyzed. Both pathogenic L. interrogans and saprophytic L. biflexa showed improved motility in highly viscous liquids, which is in agreement with previous reports on L. biflexa . Schneider and Doetsch showed that swimming speeds of externally flagellated bacteria increase to certain values with increased viscosity, but that the speeds gradually decrease in higher viscosity . In contrast, it has been reported that spirochete species translate more actively the more viscous the environment [6, 10, 11]. The morphological characteristics of spirochete species and the microscopic structures of viscous liquids are thought to allow these species to move efficiently in high viscosity .
Interestingly, L. interrogans maintains vigorous motility near physiological osmotic conditions, whereas the motility of L. biflexa is severely impaired, which suggests that pathogenic Leptospira have an advantage during migration in host animals. Certain cytoplasmic membrane proteins are believed to play a role in osmoregulation in L. interrogans . Characteristics of functional molecules such as surface proteins or lipopolysaccharides also differ between pathogenic and saprophytic Leptospira species , which may explain the adaptability of L. interrogans to physiological osmolarity. It is known that the host complement-mediated killing system is effective against L. biflexa but not L. interrogans , and that saprophytic Leptospira species are more susceptible to H2O2 than are pathogenic species . Differences in susceptibility to external stimuli appear to be critical factors that enhance virulence. Bulach et al. examined the ability of L. interrogans and Leptospira borgpetersenii serovar Hardjo, which is also a pathogenic Leptospira species, to survive and showed that >90% of L. borgpetersenii cells lost viability within 48 hr in water, whereas 100% of L. interrogans cells remained viable during the assay . The different ability to survive in low nutrient environments is likely to be related to these organisms' infection processes: L. interrogans cells are indirectly transmitted to hosts via contaminated water whereas L. borgpetersenii cells are transmitted by direct contact between bacteria and contaminated body fluids . Likewise, resistance to change in osmolarity may confer survival ability on L. interrogans during transmission. L. interrogans has a chemotactic response to hemoglobin, which suggests that it may sense hemoglobin seeping from cuts or abrasions on an animal's skin . After their initial attachment to skin, active motility retained in physiological osmolarity may facilitate the colonization of specific organs by allowing crossing of tissue barriers or mucus layers, or via other pathways such as the bloodstream.
We thank T. Masuzawa (Chiba Institute of Science, Chiba, Japan) for his kind gift of L. biflexa strain Patoc I, N. Koizumi (National Institute of Infectious Diseases, Tokyo, Japan) for his kind gift of L. interrogans serovar manilae strain UP-MMC-NIID and critical reading of the manuscript and N. Kikuchi (Rakuno Gakuen University, Hokkaido, Japan) for helpful comments.
No authors have any conflicts of interest to declare.