Surface motility of polycyclic aromatic hydrocarbon (PAH)-degrading mycobacteria

Research in Microbiology 159 (2008) 255e262 www.elsevier.com/locate/resmic

Surface motility of polycyclic aromatic hydrocarbon (PAH)-degrading mycobacteria Line Fredslund a,b, Kristel Sniegowski a, Lukas Y. Wick c, Carsten S. Jacobsen b, Rene´ De Mot d, Dirk Springael a,* a

Division of Soil and Water Management, Catholic University of Leuven, Kasteelpark Arenberg 20, B-3001 Heverlee, Belgium b Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen, Denmark c UFZ Helmholtz-Centre for Environmental Research, Department of Environmental Microbiology, Permoserstr. 15, 04318 Leipzig, Germany d Centre of Microbial and Plant Genetics (CMPG), Catholic University of Leuven, Kasteelpark Arenberg 20, B-3001 Heverlee, Belgium Received 25 October 2007; accepted 28 February 2008 Available online 19 March 2008

Abstract Surface motility of the polycyclic aromatic hydrocarbon (PAH)-degrading Mycobacterium gilvum VM552 was tested on agar and agarose plates prepared with varying amounts of gelling agents in the presence and absence of phenanthrene. Extensive spreading, originating from the point of inoculation, was observed on the surfaces of plates prepared with up to 0.3% agar and up to 0.6% agarose. The spreading velocities were 15.8 mm d1 on 0.3% agar and 19.5 mm d1 on 0.3% agarose plates. No evidence was found of accelerated or directed surface motility towards PAH crystals. The morphology of spreading M. gilvum VM552 colonies depended on both the carbon source and the type and concentration of the gelling agent. In 0.3% agar plates, M. gilvum VM552 cells were organized in 1e2-mm-wide branches of 1e5 cm length, while on agarose they slid as a homogenous monolayer across the surface. Microscopic inspection of the colonies on agar surfaces suggested that formation of branches was the combined effect of: (i) cell division and growth at the tip of a branch; (ii) propulsion of cells from the mature basal parts of a branch towards the tip; and (iii) physiologically induced reduced friction between cells and agar. Similar surface migration patterns were observed for the anthracene-degrading M. frederiksbergense LB501T. Ó 2008 Elsevier Masson SAS. All rights reserved. Keywords: Mycobacteria; Surface motility; Polycyclic aromatic hydrocarbons

1. Introduction Microbial degradation is a major factor in the breakdown of polycyclic aromatic hydrocarbons (PAHs) in soils [3], despite the low aqueous solubility and bioavailability of these pollutants. PAH-degrading bacteria display several physiological adaptations to overcome the limitation of low bioavailability. These include close attachment to the PAH source, high-affinity uptake systems, production of biosurfactants [9] and active movement along continuous liquid films covering fungal hyphae [11]. In addition, various PAH-degrading bacteria have been shown to display positive chemotaxis towards low* Corresponding author. Tel.: þ32 16 32 16 04; fax: þ32 16 32 19 97. E-mail address: [email protected] (D. Springael). 0923-2508/$ - see front matter Ó 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2008.02.007

molecular-weight PAHs such as naphthalene and phenanthrene [12,14,15]. Migration and subsequent attachment to PAH-contaminated soil particles will increase bioavailability of the PAH by increasing the flux of PAH to the cells. PAH-directed motility has been studied only for flagellated PAH-degrading bacteria such as Pseudomonas, despite the observation that many PAH degraders derived from long-term PAH-contaminated soils are non-flagellated Gram-positive bacteria such as Mycobacterium spp. The latter species are often present in high numbers in PAH-contaminated soil and are thus considered to play an important role in the in situ degradation of PAHs [9,20]. Unlike Pseudomonas, several environmental Mycobacterium species are able to use three- and four-ring PAHs, such as phenanthrene, pyrene and fluoranthene, as sole sources of carbon and energy, and to co-metabolically

256

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

degrade benzo[a]pyrene [4]. PAH-degrading mycobacteria display various physiological adaptations for survival in PAHcontaminated soils, such as high specific affinities for PAH [8,20,22,23] or biofilm formation on PAH-containing surfaces [8,20,22,23]. Although mycobacteria were previously thought to be non-motile, Martı´nez et al. (1999) published evidence that Mycobacterium isolates such as the fast-growing saprophytic Mycobacterium smegmatis mc2155 and the slow-growing opportunistic pathogen M. avium can spread on the surface of motility agar plates by a sliding mechanism as defined by Henrichsen [6]. M. smegmatis spreads on agarose as a monolayer of cells with close cell-to-cell contact and its spread is accompanied by growth and division of the individual cells. Martı´nez et al. (1999) stated that this form of motility is likely to play a significant role in the ability of mycobacteria to colonize surfaces in the environment and the host. The purpose of this study was to examine whether PAHdegrading environmental isolates of Mycobacterium also possess surface-associated motility and whether this movement is influenced by the presence of PAH in the surrounding environment. Therefore, cells of phenanthrene-degrading Mycobacterium gilvum VM552 and, where appropriate, of anthracene-degrading Mycobacterium frederiksbergense LB501T were inoculated into the center of phosphate-buffered minimal medium (MM) agar or agarose motility plates in the presence or absence of crystalline PAH (phenanthrene or anthracene). The effects of pre-experimental growth conditions, surface moisture and supplementary carbon source on the velocity and pattern formation of the spreading cells were investigated. 2. Materials and methods 2.1. Bacterial strains and culture conditions M. gilvum VM552 was isolated from PAH-contaminated soil sampled from a bitumen factory. This strain can use phenanthrene and pyrene as its sole sources of carbon and energy (D. Springael, unpublished observations). M. frederiksbergense LB501T was isolated from a mixture of PAH-contaminated soils sampled from various gas manufacturing plants, and it utilizes anthracene as its sole source of carbon and energy [2]. Both strains were routinely grown at 25  C in liquid phosphate-buffered minimal medium (MM) (pH 7.5) [20] modified either with the addition of 1 mM D-glucose or with phenanthrene (VM552) or anthracene (LB501T) in crystalline form, as sole carbon and energy sources. 2.2. Surface spreading assays Motility tests were performed in soft agar or agarose plates as described previously [13]. Twenty ml of sterile medium was solidified in 9 cm-diameter PVC plastic Petri dishes (Greiner Bio-One NV) with agar (Select Agar, Nusieve) or agarose (Cambrex Bio Science) at concentrations of 0.3%, 0.6%, 0.9%, 1.2% or 1.5%. Two different media were used for motility tests: minimal media (MM) and MM amended with

1 mM D-glucose. In experiments with strain VM552, when appropriate phenanthrene was added on top of the agar or agarose surface in three different modes. In a first mode, phenanthrene was added as a distant PAH source as follows: the center of the agar surface was covered with a sterile 25 mm diameter 0.45-mm filter (MFÔ-Membrane filters, Millipore) or 25 mm diameter parafilm cover prior to spreading phenanthrene crystals over the agar/agarose surface. By removal of the filter or parafilm, a fixed distance of a minimum of 12.5 mm was created between the inoculation point at the center of a phenanthrene crystal-free zone and phenanthrene crystals around its perimeter. After addition of phenanthrene, plates were left for 3 days at 25  C prior to inoculation to allow phenanthrene to dissolve and diffuse in the agar/agarose. This allowed for phenanthrene to dissolve in agar/agarose. The concentration of phenanthrene in the agar at the center of the plate was 0.623 (SD 0.053) mg l1 at the time of inoculation as shown by high performance liquid chromatography (HPLC) analysis of a 1 cm2 sample of agar taken from the center of a phenanthrene-modified agar plate (the maximum aqueous solubility of phenanthrene is 1.2 mg l1). The sample was weighed and left for 0.5 h in 1.5 ml hexane (HPLC grade, Biosolve LTD) after brief vortex mixing. The concentration of PAH in the hexane phase was measured on a HPLC apparatus (Lachrom, Merck Hitachi) equipped with a C-18 column LiChroCART 125-4, a Purospher STAR RP18 end-capped column (125 mm in length; 4 mm in diameter) and a UV-VIS L-7420 detector (Lachrom, Merck Hitachi). A control of the filter hollow set-up was prepared with application of sterile filters to the surface of the agar, without addition of phenanthrene crystals around the filter hollow. In a second mode, in order to study unidirectional spreading of the bacteria, a line of phenanthrene crystals was deposited at one side of the center of the plate at around 1 cm distance from the border of the Petri dish. In a third mode, phenanthrene was uniformly spread as a thin opaque layer over the agar surface using 3 ml of a 0.2% phenanthrene solution in acetone as described previously [10]. In all cases, bacteria were inoculated at the center of the Petri dishes by pipetting 5 ml of a cell suspension obtained from a liquid culture grown in minimal medium containing phenanthrene or glucose until an optical density at 600 nm of 0.4. Prior to inoculation, the culture optical density at 600 nm was adjusted to 0.02. All test conditions were run in triplicate. After inoculation, plates were sealed with parafilm and incubated at 25  C in a dark chamber. Spreading was evaluated visually each day and migration of bacteria from the center of the Petri dish measured on days 2, 3, 4, 5, 6, 7, 8, 10, 12, and 17. Analogous experiments following the second mode were performed with M. fredriksbergense LB501T and anthracene crystals as distant carbon source covering either minimal medium agar (0.3%) or agarose (0.3%) plates. In contrast to strain VM552, however, strain 501T was inoculated with the tip of a sterile tooth pick containing about ca. 107 cells obtained from colonies growing either on LB agar or minimal medium agar with anthracene as sole carbon source.

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

2.3. Phase-contrast microscopy Cells on the surface of growth medium were visualized with an Olympus BX41 inverted microscope at 200 and 400 magnifications and an Olympus BX51 upright microscope. The images were captured with an Olympus DP50CU 3.0 camera integrated with a PC using analySIS Docu software (Version 3.2) produced by Soft Imaging System GmbH (http://www.soft-imaging.net). Images were processed with Microsoft Powerpoint and Photoshop 4.0.1 (Adobe) software packages. 3. Results 3.1. Surface motility of M. gilvum VM552 on agar plates M. gilvum VM552 was found to randomly colonize the surfaces of 0.3% MM motility agar plates independently of the presence of amended glucose or phenanthrene (Fig. 1). Spreading of M. gilvum VM552 was restricted to the surface of the agar. A lag-phase of 3e4 days generally preceded spreading. During that time, cells multiplied and built a multilayered dense colony around the point of inoculation before starting to produce fine, radially spreading threads in all directions. On plates containing a central zone encircled by PAH-crystals, VM552 first colonized the point of inoculation, created a fine web of closely interspersed branches in the crystal-free zone (Fig. 1, picture C), and then changed its morphology at the fringe of the zone to produce the larger distinct finger-like branches observed on the other plates (Fig. 1, pictures C and G). However, the same spreading features were observed on similar treated control plates where no phenanthrene crystals were added indicating that this behavior relates to topical changes of the agar surface by placing and removing the filter or parafilm which was used to create the zone. Maximum spreading velocities of the colonies were around 7.0 mm d1 (Table 1). Although the presence of phenanthrene on the MM motility agar plates did not affect this maximum spreading velocity (Table 1), it was reached earlier when phenanthrene was present than when it was absent (Fig. 2A). During the incubation period, both the number and width of the branches increased and individual branches divided into multiple finger-like extensions. Interestingly, branches stopped growing or changed direction when they met other branches on the agar surface. In cases where distant phenanthrene crystals were present, M. gilvum VM552 created a fine network of threads around each phenanthrene crystal (Fig. 1, pictures C and G) and colonized the surface of the crystals with a multilayer of cells of distinctive wet mucoid appearance (Fig. 1, pictures D and H). In plates with the thin phenanthrene spray layer, the strain formed a clearing zone in the phenanthrene film around the colony and then started to spread as branches, creating clearing zones along their borders (Fig. 1, picture B). Addition of glucose to MM motility 0.3% agar medium slightly increased the spreading velocity of M. gilvum VM552 (Table 1). However, no differences in maximum spreading rates and distances were observed between plates

257

containing glucose with and without phenanthrene (Fig. 1, pictures E and G; Table 1). An increase in the agar concentration from 0.3% to 0.6% completely inhibited the spread of M. gilvum VM552 across the agar surface (Table 1). In contrast, the nature of the carbon source during pre-experimental growth conditions had no influence on colonization of the agar surface (Table 1). When phenanthrene crystals were added at one side of the inoculation point to test unidirectional spreading towards phenanthrene crystals, strain VM552 spread out randomly with fingers over the agar surface in all directions. This response was observed on all plates (with and without glucose). In the neigborhood of the crystals, VM552 seemed to produce a finer web and more biomass, finally resulting in colonization of the phenanthrene crystals (data not shown). However, on the macroscopic scale, there was no evidence that strain VM552 spread preferentially towards the phenanthrene source since, in some plates (both with and without glucose) the branches of VM552 spread away from and never reached the phenanthrene crystals (data not shown). 3.2. Surface motility of M. gilvum VM552 on agarose plates Substitution of agar with agarose as a gelling agent significantly increased both maximal spreading velocity and spreading distances after 8 days (Fig. 2B and Table 1). On 0.6% agarose MM plates, M. gilvum spread at a maximal velocity of approximately 3.5 mm days1. Small spreading colonies with a radius of 6e7 mm were also found on MM plates with an agarose concentration of up to 0.9%. Similarly to spreading on agar plates, spreading morphology and velocity depended on the concentration of agarose as well as the carbon source. As with the agar plates, branches up to 10 mm in width were formed on 0.3% agarose plates. These branches were often interlinked by a transparent monolayer of homogeneously spreading cells over the agarose surface (data not shown). In 0.6% agarose in MM, M. gilvum VM552 colonized the agarose surface as one continuous monolayer of cells spreading radially outwards (data not shown). In 0.3% agarose in the presence of phenanthrene crystals, the spreading branches of M. gilvum VM552 cleaved or sank into the surface of the agar (data not shown). As observed on agar surfaces, the growth substrate of the inoculum did not affect the spreading behavior of strain VM552 (Table 1). 3.3. Microscopic inspection of spreading morphology of M. gilvum VM552 To obtain further information on the surface-related motility of M. gilvum VM552, colonization of a MM 0.3% agar surface in the presence of glucose and distant phenanthrene crystals was examined by inverted microscopy analysis. The pattern of spreading and the cellular organization of VM552 on the agar surface were clearly different within the area initially occupied by the filter compared with cells outside the area (Fig. 3, picture A).

Fig. 1. Images of the characteristic morphology of M. gilvum VM552 spreading on the surface of 0.3% motility agar plates with different media. A, minimal medium (MM) without C-source; B, MM with phenanthrene film layer; C, MM with distant phenanthrene crystals; D, enlargement of C with black arrows pointing to phenanthrene crystals overgrown by M. gilvum VM552 and white arrow pointing to a non-colonized phenanthrene crystal; E, MM with 0.1 mM glucose; F, MM with 0.1 mM glucose and phenathrene film layer; G, MM with 0.1 mM glucose and distant phenanthrene crystals; H, enlargement of G with black arrows pointing to phenanthrene crystals overgrown by M. gilvum VM552 and white arrows pointing to non-colonized phenanthrene crystals. The plates were incubated at 25  C for 10 days. In plates B and F, the remaining phenanthrene spray layer can be seen as an opaque white layer at the upper and left side of the plates.

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

259

Table 1 Spreading velocity and spreading distance of M. gilvum VM552 on motility agars differing in gelling agents, gelling agent concentration and carbon source Pre-culturing C-source

Gelling agent in Petri dish

C-source in minimal medium

Max. spreading velocity (mm d1)a

Distance at 8 days (mm)a,b

Phen Phen Phen Phen Phen Phen Phen Phen Phen Glu Glu Glu Phen Phen Phen Phen Phen Phen Glu Glu Glu

0.3% 0.3% 0.3% 0.6% 0.6% 0.6% 0.3% 0.3% 0.3% 0.3% 0.3% 0.3% 0.3% 0.3% 0.3% 0.6% 0.6% 0.6% 0.3% 0.3% 0.3%

e Phen Distant Phen Film e Phen Distant Phen Film MM þ Glu Glu þ Phen Distant Glu þ Phen Film e Phen Distant Phen Film e Phen Distant Phen Film e Phen Distant Phen Film e Phen Distant Phen Film

6.4  2.8 7.0  3.8 9.5  4.0 0.3  0.3 0.3  0.6 0.3  0.3 11.7  3.3 15.0  4.2 12.5  4.4 5.7  3.3 6.0  3.5 15.8  1.8 14.2  3.8 16.2  4.0 15.5  4.4 3.3  0.3 4.0  0.9 0.8  0.6 16.3  2.0 19.5  7.9 18.2  2.3

5.0  1.3 19.5  0.9 28.8  2.0 2.0  0.0 2.7  1.2 2.3  0.3 34.8  3.2 37.0  0.7 35.2  3.8 12.0  4.8 24.5  6.5 42.0  0.0 35.3  5.0 42.0  0.0 42.0  0.0 12.3  0.6 13.7  0.6 3.8  2.0 39.7  4.0 42.0  0.0 42.0  0.0

agar agar agar agar agar agar agar agar agar agar agar agar agarose agarose agarose agarose agarose agarose agarose agarose agarose

a

Values are means  standard deviations from three assays. 42.0 mm is the maximal radius of spreading in a 9 cm Petri dish. Abbreviations: Phen, phenanthrene; Glu, glucose; Phen Distant, distant phenanthrene; Phen Film, phenanthrene uniformly spread as a thin opaque film. b

Cells forming a branch were organized as monolayers and/ or multilayered structures. Monolayers of cells formed the borders of all branches. In mature branches, the flanking monolayers were organized as small loops which resembled the fine hairs of a growing plant root (Fig. 3, picture B). Within the branches, cells were organized as multiple layers of cells with a moist inner core in which individual cells were observed to be transported towards the tip of the branch along a liquid ‘‘highway’’. The cells appeared to be passively transported by a flow of fluid in the inner core of the branch towards the tip (Fig. 3B). At the growing tip of the branch, cells were both loosely packed in a monolayer and oriented along the longitudinal axis of the growing branch. Monolayers normally developed over time into a multilayered structure of loosely packed cells with mixed orientation, by which time the branch had stopped growing (Fig. 3, pictures C and D). 3.4. Surface-associated motility of M. frederiksbergense LB501T In order to assess whether surface-related motility is a general trait of PAH-degrading mycobacteria, a series of motility experiments was performed with the anthracene-degrading Mycobacterium frederiksbergense LB501T. Strain LB501T showed very similar behavior to strain VM552 on agar and agarose plates. M. frederiksbergense LB501T developed five to ten distinct finger-like extensions on all media with and without anthracene. The extensions were randomly distributed and covered the whole agar surface within seven days with an apparent maximum spreading velocity of >6 mm day1 (data not shown). Neither an increased rate of spreading nor unidirectional growth towards distant anthracene could be observed

when compared with media lacking anthracene (data not shown). Direct contact of dispersed LB501T bacteria with solid anthracene resulted in biofilm formation on the anthracene surface, as well as the formation of finely branched ‘subfingers’ spreading further over the field of anthracene-covered agar surfaces. 4. Discussion In this study, we provide the first evidence that PAH-degrading mycobacteria are able to spread across surfaces. Overall, the macroscopic spreading behavior of strains M. gilvum VM552 and M. frederiksbergense LB501T on agar and agarose surfaces shares several features with the surface sliding motility of M. smegmatis mc2155. These include: (i) the formation of finger-like colony extensions on agar surfaces; (ii) increased spreading on agarose- compared with agar-containing media; (iii) similar motility rates; (iv) similar effects of moisture content on surface-related motility; and (v) the fact that spreading on a solid surface is accompanied by growth [13]. For M. smegmatis mc2155 and M. avium, sliding motility was found to depend on a combination of both specific cellto-cell interactions and growth, as increased spreading velocities were observed in the presence of an additional carbon source [13]. The latter may also explain the differences found for strain VM552 in spreading on MM media and MM media containing phenanthrene and/or glucose as additional sources of carbon and energy. Glucose is a carbon source for M. gilvum VM552 and simultaneous utilization of glucose and PAH has been described for several PAH-degrading Mycobacterium strains ([24], L. Bastiaens and D. Springael, unpublished results). Strain VM552 also performed surface sliding

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

260

Colony diameter (mm)

A

MM with 0.3% agar 100 MM MM + Phe crystals MM + Phe spray MM + Filter control

80

60

40

20

0

0

2

4

6

8

10

12

14

16

18

14

16

18

Days

Colony diameter (mm)

B

MM with 0.3% agarose 100

MM MM + Phe cryst MM + Phe spray MM + Filter con

80

60

40

20

0

0

2

4

6

8

10

12

Days Fig. 2. Time-dependent colony diameter spreading of M. gilvum VM552 on MM 0.3% motility agar and agarose plates. (A) Mean triplicate colony diameters (mm) of M. gilvum VM552 in 0.3% agar with minimal medium (MM) eCe, MM with distant phenathrene crystals eBe, MM with phenanthrene spray layer e;e, and MM with filter control e7e. (B) Mean triplicate colony diameters of M. gilvum VM552 in 0.3% agarose with minimal medium (MM) eCe, MM with distant phenathrene crystals eBe, MM with phenanthrene spray layer e;e, and MM with filter control e7e.

motility in the absence of an additional carbon source on MM agar. Similar results have been reported by Martı´nez and coworkers for M. smegmatis mc2155. As reported for strain mc2155, growth on MM without added C-source by VM552 was unlikely to be supported by nutrients that were carried over from the growth medium, since colonies originating from unwashed and washed cells behaved identically (data not shown). Strain VM552 may either have utilized low concentrations of carbon sources present as chemical impurities in the agar or have accumulated internal storage compounds in order to support cell growth in the case of external carbon depletion [1,21]. Spreading of M. gilvum was observed on MM agarose plates without additional C-source, indicating that strain VM552 was using storage compounds for observed growth and spreading. As for M. smegmatis mc2155, surface-related spreading of M. gilvum VM552 appears to be a group event, performed in

the form of coordinated growth of the cells. Microscopic inspection of the spreading behavior of strain VM552 on agar seems to be the result of a combination of several phenomena, including: (i) cell division and growth at the tip of a branch; (ii) propulsion of additional cells from the mature basal parts of a branch towards the tip; and (iii) reduced friction between cells and agar, as proposed by Martı´nez et al. [13,21]. In M. smegmatis mc2155 and M. avium, the ability to spread over the surface has been related to the presence of glycopeptidolipids (GPLs). Non-sliding M. smegmatis mc2155 mutants are defective in genes involved in GPL biosynthesis [16,17]. This is a Mycobacterium-specific class of amphiphilic molecules located in the outermost layer of the cell envelope which are believed to reduce the friction between cells and the surface [13]. Also, in Mycobacterium marinum, a group of antigenic GPLs, lipo-oligosaccharides were shown to play an important role in sliding motility [18], while the amount and composition of GPLs in the cell wall has been identified as a possible determinant of sliding motility of opportunistic pathogenic Mycobacterium abscessus [7]. For strain M. smegmatis mc2155, microscopic observations of sliding were focused on cells spreading on 0.3% agarose in which the colony was expanding homogeneously in all directions. Spreading occurred as a monolayer of cells arranged in pseudofilaments with close cell-to-cell contact in cases where the experiments were performed in MM solidified with agarose and without an added carbon source [13]. However, the pattern of spread of cells on media containing additional carbon sources became more complex and showed cycles of spread followed by conversion of the monolayer into a dense cell mass [13]. The spreading behavior of M. smegmatis was characterized by a sliding mechanism [13] as defined by Henrichsen [5]. Our microscopy observations of the multilayered organization of M. gilvum cells as finger-like branches suggest that the motility phenomenon of M. gilvum is best described as a morphological variant of Henrichsen’s sliding mechanism [5,6], because expanding colonies of M. gilvum VM552 also displayed multilayered branches, and single cells present in a liquid core region of the multilayered branch seemed to be transported towards the tip. Video imaging of transport of individual cells in the core region of the branches was unfortunately not feasible. The transport of cells was, however, clearly observed under the microscope and time series photo imaging of the phenomenon. From this series of images, the speed of single cell transport was estimated to be on the order of 10e20 mm per min. Transport of cells in the core region hence appeared much faster than spreading of single cells at the growing tips of branches. Under the experimental conditions used here, we did not acquire any evidence that spreading could be PAH-directed for either strain VM552 or LB501T. No significant differences in maximum spreading velocities were found between MM plates with and without phenanthrene. The maximum spreading velocity was reached earlier on plates with phenanthrene, but we attribute this to the additional growth of the bacteria on phenanthrene rather than to accelerated spreading induced by phenanthrene. Furthermore, in the presence of glucose,

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

261

Fig. 3. Merged micrographs of a M. gilvum VM552 colony after 17 days of spreading on the surface of a 0.3% agar plate containing D-glucose in the presence of phenanthrene crystals. (A) Phase-contrast micrographs of M. gilvum VM552 (magnification: 400; bar size: 50 mm). The direction of spread is from right to left. The images show the edge of the filter zone within which VM552 produced a fine network of branches (right). From the edge of the finely branched network towards the phenanthrene crystals surrounding the filter zone, VM552 developed a coarse network of thicker branches that spread from one phenanthrene crystal to another. The black arrows indicate small phenanthrene crystals. (B) Phase-contrast micrographs of M. gilvum VM552 (magnification: 1000; bar size: 20 mm). The direction of spread is from left to right. The image shows colonized phenanthrene crystals (black arrows), from which the VM552 spreads in organized branches away from the expanding colony. A well-developed branch is entering the image at the top and leaves the picture to the right. At the base of a growing branch it consists of flanking monolayers of cells organized in loops that resemble fine root hairs (encircled). In the core of the branch the cells are organized entirely differently in a multilayer of cells with a moist inner core, where individual cells seem to be transported towards the tip of the branch in a liquid ‘‘highway’’ (white arrow). (C) Micrograph of a growing tip of M. gilvum VM552 passing close to a phenanthrene crystal at 2000 magnification (bar size: 5 mm). The growing tip is a monolayer of loosely packed cells (bottom left). At the top (right) the young branch is developing into a multilayered dense structure. (D) Micrograph of a young branch at 10,000 magnification. (Bar size: 1 mm). The young branch is developing from a monolayer in which individual cells can be easily recognized, into a multilayered organization of cells appearing as scattered white areas in the cell mass.

maximum spreading velocities and distances reached after 8 days of incubation were very similar on plates with and without phenanthrene. On the other hand, clustering of M. gilvum VM552 colony branches around and over the phenanthrene crystals (as shown in Fig. 1) suggests that there may be some sort of directional growth, at least at close quarters. However, we found no evidence for unidirectional growth in plates supplied with distant phenanthrene crystals at one side of the inoculation point. Furthermore, it is difficult to distinguish between directed growth at close quarters and increased

growth due to the fact that the C-source is nearby, resulting in a higher flux of phenanthrene to the bacteria. Growth will most likely occur at the site at which the carbon source is located and will result in a larger build-up of biomass (Fig. 1). Our data have important ecological significance. Although no evidence was found that the presence of phenanthrene has a direct effect on spreading patterns, we show that the ability to spread across surfaces can be used by both PAH-degrading Mycobacterium strains to reach distantly located PAH sources. In polluted oils, PAH tend to sorb soil constituents due to their

262

L. Fredslund et al. / Research in Microbiology 159 (2008) 255e262

low aqueous solubility and are heterogeneously distributed on the macro- and micro-scale. The distance between bacteria and pollutant is seen as a major obstacle for decontamination [9]. Sliding motility can contribute to the survival and proliferation of PAH-degrading mycobacteria in soil by allowing them to explore and colonize free surfaces eventually containing remote PAH sources. Subsequent attachment to the PAHcontaining surfaces might increase bioavailability of the PAH by increasing the flux of PAH to the cells. In that context, we have recently shown that, in long-term PAH-contaminated soil, PAH-degrading mycobacteria are primarily associated with PAH-enriched soil components [19]. Moreover, the tested strains belong to two different branches of Mycobacterium phylogeny, indicating that the capacity of surface sliding motility is widely distributed within Mycobacterium and among phylogenetically diverging PAH-degrading strains. Further research in our laboratory is focussing on the mechanism of surface sliding motility in PAH-degrading mycobacteria and the role of surface motility in colonization and PAH degradation in contaminated soil. Acknowledgements This work was financed by FWO-Flanders (FWO grant no. G.0254.03), the Danish Research Councils (SOUND grant no. 23-02-0152) and was supported by Novozymes A/S, Bagsværd, Denmark. We thank P. Declerck for assistance in inverted microscopy analysis and S. Arnold, J. Reichenbach, B. Wu¨rz and R. Remer for practical help with strain M. frederiksbergense LB501T in the laboratory at UFZ.

[8]

[9] [10]

[11]

[12]

[13] [14]

[15]

[16]

[17]

[18]

[19]

References [1] Alvarez, H.M., Steinbuchel, A. (2002) Triacylglycerols in prokaryotic microorganisms. Appl. Microbiol. Biotechnol. 60, 367e376. [2] Bastiaens, L., Springael, D., Wattiau, P., Harms, H., deWachter, R., Verachtert, H., et al. (2000) Isolation of adherent polycyclic aromatic hydrocarbon (PAH)-degrading bacteria using PAH-sorbing carriers. Appl. Environ. Microbiol. 66, 1834e1843. [3] Cerniglia, C.E. (1992) Biodegradation of polycyclic aromatic hydrocarbons. Biodegradation 3, 351e368. [4] Cheung, P.Y., Kinkle, B.K. (2001) Mycobacterium diversity and pyrene mineralization in petroleum-contaminated soils. Appl. Environ. Microbiol. 67, 2222e2229. [5] Harshey, R.M. (2003) Bacterial motility on a surface: many ways to a common goal. Ann. Rev. Microbiol. 57, 249e273. [6] Henrichsen, J. (1972) Bacterial surface translocation e survey and a Classification. Bacteriol. Rev. 36, 478e503. [7] Howard, S.T., Rhoades, E., Recht, J., Alsup, A., Kolter, R., Lyons, C.R., et al. (2006) Spontaneous reversion of Mycobacterium abscessus from a smooth to a rough morphotype is associated with reduced expression

[20]

[21]

[22]

[23]

[24]

of glycopeptidolipid and reacquisition of an invasive phenotype. Microbiology 152, 1581e1590. Johnsen, A.R., Karlson, U. (2004) Evaluation of bacterial strategies to promote the bioavailability of polycyclic aromatic hydrocarbons. Appl. Microbiol. Biotechnol. 63, 452e459. Johnsen, A.R., Wick, L.Y., Harms, H. (2005) Principles of microbial PAH-degradation in soil. Environ. Pollut. 133, 71e84. Kiyohara, H., Nagao, K., Yana, K. (1982) Rapid screen for bacteria degrading water-insoluble, solid hydrocarbons on agar plates. Appl. Environ. Microbiol. 43, 454e457. Kohlmeier, S., Smits, T.H.M., Ford, R.M., Keel, C., Harms, H., Wick, L.Y. (2005) Taking the fungal highway: mobilization of pollutant-degrading bacteria by fungi. Environ. Sci. Technol. 39, 4640e4646. Law, A.M.J., Aitken, M.D. (2003) Bacterial chemotaxis to naphthalene desorbing from a nonaqueous liquid. Appl. Environ. Microbiol. 69, 5968e5973. Martı´nez, A., Torello, S., Kolter, R. (1999) Sliding motility in Mycobacteria. J. Bacteriol. 181, 7331e7338. Ortega-Calvo, J.J., Marchenko, A.I., Vorobyov, A.V., Borovick, R.V. (2003) Chemotaxis in polycyclic aromatic hydrocarbon-degrading bacteria isolated from coal-tar- and oil-polluted rhizospheres. FEMS Microbiol. Ecol. 44, 373e381. Parales, R.E., Harwood, C.S. (2002) Bacterial chemotaxis to pollutants and plant-derived aromatic molecules. Curr. Opin. Microbiol. 5, 266e 273. Recht, J., Martinez, A., Torello, S., Kolter, R. (2000) Genetic analysis of sliding motility in Mycobacterium smegmatis. J. Bacteriol. 182, 4348e4351. Recht, J., Kolter, R. (2001) Glycopeptidolipid acetylation affects sliding motility and biofilm formation in Mycobacterium smegmatis. J. Bacteriol. 183, 5718e5724. Ren, H., Dover, L.G., Islam, S.T., Alexander, D.C., Chen, J.M., Besra, G.S., et al. (2007) Identification of the lipooligosaccharide biosynthetic gene cluster from Mycobacterium marinum. Mol. Microbiol. 63, 1345e1359. Uyttebroek, M., Breugelmans, P., Janssen, M., Wattiau, P., Joffe, B., Karlson, U., et al. (2006) Distribution of the Mycobacterium community and polycyclic aromatic hydrocarbons (PAHs) among different size fractions of a long-term PAH-contaminated soil. Environ. Microbiol. 8, 836e847. Uyttebroek, M., Ortega-Calvo, J.J., Breugelmans, P., Springael, D. (2006) Comparison of mineralization of solid-sorbed phenanthrene by polycyclic aromatic hydrocarbon (PAH)-degrading Mycobacterium spp. and Sphingomonas spp. Appl. Microbiol. Biotechnol 72, 829e 836. Waltermann, M., Hinz, A., Robenek, H., Troyer, D., Reichelt, R., Malkus, U., et al. (2005) Mechanism of lipid-body formation in prokaryotes: how bacteria fatten up. Mol. Microbiol. 55, 750e763. Wick, L.Y., Colangelo, T., Harms, H. (2001) Kinetics of mass transferlimited bacterial growth on solid PAHs. Environ. Sci. Technol. 35, 354e361. Wick, L.Y., de Munain, A.R., Springael, D., Harms, H. (2002) Responses of Mycobacterium sp LB501T to the low bioavailability of solid anthracene. Appl. Microbiol. Biotechnol. 58, 378e385. Wick, L.Y., Wattiau, P., Harms, H. (2002) Influence of the growth substrate on the mycolic acid profiles of mycobacteria. Environ. Microbiol. 4, 612e616.