A novel method for the isolation of motile bacteria using gradient culture systems

A novel method for the isolation of motile bacteria using gradient culture systems

Journal of Microbiological Methods 46 Ž2001. 141–147 www.elsevier.comrlocaterjmicmeth A novel method for the isolation of motile bacteria using gradi...

121KB Sizes 5 Downloads 51 Views

Journal of Microbiological Methods 46 Ž2001. 141–147 www.elsevier.comrlocaterjmicmeth

A novel method for the isolation of motile bacteria using gradient culture systems Rebecca Thomson a , Roger Pickup b, Jonathan Porter a,) a

Hatherly Laboratories, School of Biological Sciences, UniÕersity of Exeter, Exeter, EX4 4PS, UK b Centre for Ecology and Hydrology, Windermere, Cumbria, UK Received 10 October 2000; received in revised form 25 March 2001; accepted 28 March 2001

Abstract Isolation of motile bacteria from stream water samples was achieved by using Lutrol F127 Žpoloxamer 407. as a gelling agent in culture media. This block copolymer has the property of repeatedly liquefying and solidifying at low and high temperatures, respectively. The ability of motile bacteria to move through liquid-state Lutrol F127 towards a higher nutrient concentration was exploited. After establishment of the nutrient gradient and inoculation, the system was cooled to liquefy the medium and kept liquid to allow motile bacteria to move. Raising the temperature allowed solidification and prevented further movement. Colonies could be easily removed. The proportion of motile isolates Ždetermined by microscopic observation. increased from 42% in the indigenous population to 100% after isolation using the gradient system. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Isolation; Motile bacteria; Nutrient gradient; Novel gelling agent

1. Introduction The majority of bacteria from natural environments have not as yet been obtained in laboratory culture and further research into basic culture methods is required. The goal of obtaining cells from the natural world in stable, pure laboratory culture is worthwhile from the view of establishing ecophysiological potential by laboratory study, as well as facilitating the development of molecular tools for direct studies. Lutrol F127 Žpoloxamer 407. is a fully autoclavable block copolymer based on polyoxyethylene ) Corresponding author. Tel.: q44-1392-264607; fax: q441392-263700. E-mail address: [email protected] ŽJ. Porter..

and polypropylene Ž70:30.. At low temperature, a Lutrol solution is liquid, but becomes solid upon warming. The exact temperature at which this occurs is dependent upon the concentration of the gelling agent. This liquefactionrsolidification process can be performed many times without apparent detriment to the polymer structure, and offers the potential to examine bacterial behaviour in a liquid medium before solidification to trap cells and hold them in position prior to further analysis or manipulation. Use of this co-polymer in bacteriological media was examined by Gardener and Jones Ž1985. as a gelling agent in general purpose growth media for freshwater bacteria. No inhibition of bacterial growth was reported, and the clarity of the medium was found to be of benefit when examining plates for small, translucent colonies. Gardener and Jones Ž1985. also

0167-7012r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 7 0 1 2 Ž 0 1 . 0 0 2 5 8 - 5

142

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

noted that it might be beneficial to be able to produce pour plates without introducing a heat shock to cells taken from an environmental sample. This copolymer has also been investigated as a method to deliver cells onto solid supports to model cell behaviour in biofilms ŽGilbert et al., 1998., and to facilitate delivery of cells onto solid supports for subsequent biofilm formation ŽWirtanen et al., 1998.. Motility is widely distributed in the microbial world and is a property of many pathogenic bacteria. Targeted isolation of motile cells is of general interest for studies using pathogenic bacteria, e.g. aeromonads that are pathogenic to fish ŽMerino et al., 1997. and studies on subgingival dental plaque ŽOjima et al., 1998.. Various methods have been proposed for the isolation of motile bacteria, e.g. sliced agar media ŽOlson, 1996. and the modified semi-solid Rappaport Vassiliadis medium ŽPoppe and Duncan, 1996. for isolation of motile salmonellae. Furthermore, the use of gradients within agar media has also been investigated to assess the response of bacterial cultures to simultaneous variations in solute composition and concentration Že.g. Peters et al., 1991.. This approach has allowed determination of the limits of tolerance of environmental factors for several species. In this work, a novel approach was used to develop a generic protocol for the isolation of motile heterotrophic bacteria from freshwater stream samples, using the reversible liquefactionrsolidification properties of Lutrol F127 to create gradients and allow bacterial movement in response to those gradients. Further developments could investigate the response of known motile bacteria to a choice of carbon sources, allowing investigations into which carbon sources the isolates preferred to utilise. This may ultimately lead to a more rational approach to the design of growth media.

2. Materials and methods 2.1. EnÕironmental sampling Water samples Ž20 ml. were collected in sterile bottles from Taddiforde Brook, ŽGreat Britain Ordnance Survey grid reference SX 920 938.. This stream receives waters from the University of Exeter

farmland and botanic gardens. One-milliliter samples were concentrated Žconcentrated stream water, CSW. by centrifugation at 11 600 = g; 90% of the original volume was carefully removed, keeping the pipette tip just below the liquid surface, to avoid disturbing the pellet, before addition of 10% volume sterile water to a final volume of 200 ml. The pellet was resuspended by vortexing. CSW was used to inoculate the gradient media. Sample processing was always initiated within 20 min of collection from the stream. 2.2. Bacterial growth conditions For all media, the nutrient source used was R2 ŽReasoner and Geldreich, 1985., either at full strength Ž=1. or diluted to 1r10 Ž=0.1. or 1r100 Ž=0.01. strength. R2 agar ŽR2A. was prepared as described ŽReasoner and Geldreich, 1985. and R2 broth ŽR2B. was R2A with agar omitted. Lutrol F127 was obtained from BASF ŽBASF, Colours and Chemicals Division, Cheadle, UK.. Stock Lutrol F127 solutions were prepared by the addition of 10 gr100 ml Lutrol F127 powder each day into distilled water held at 58C until a 50 gr100 ml solution was achieved. The solution was then held at 58C for a further 24 h to ensure complete dissolution, prior to autoclaving. Stock Lutrol F127 solutions were then cooled to room temperature after autoclaving, and chilled to 58C to liquefy. R2-supplemented Lutrol F127 ŽR2L. was prepared by diluting a 50 gr100 ml Lutrol F127 solution with concentrated R2B and sterile distilled water, to give a medium containing 30 gr100 ml Lutrol F127 with standard R2 nutrient, referred to as 1 = R2L. This stock was used to prepare 0.1 = R2L and 0.01 = R2L in 30 gr100 ml Lutrol F127. A calibration curve of Lutrol concentration vs. gelling temperature was also prepared. 2.3. Preparation of nutrient gradient culture systems The reversible liquefactionrsolidification properties of Lutrol F127 were used to create nutrient gradients and allow bacterial movement in response to those gradients. Vertical gradients were prepared using purpose-designed apparatus and using standard laboratory bottles. Horizontal gradients were prepared using standard culture vessels.

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

2.3.1. Preparation of Õertical nutrient gradient culture systems Purpose-built, vertical perspex plates Ž98-mm internal width= 90-mm internal height= 6-mm internal depth, plus 1-mm-thick butyl rubber gasket, total vessel volume available 70 ml; Fig. 1. were used to prepare initial nutrient gradient systems. Repeated autoclaving of the assembled plates led to warping, thus each component of the plates was alcohol-sterilised and rinsed with sterile water prior to assembly. This was repeated after assembly and the whole apparatus was equilibrated at 248C. Gradients were constructed in one of two ways, with either high nutrient levels at the base of the apparatus Žand low nutrient levels at the top. or vice versa. Both gradient systems were inoculated in the lowest nutrient part of the gradient. Inoculation was achieved by pipetting 400-ml CSW Žequivalent of 2-ml original sample. into the bottom of the apparatus or on top of the final layer of medium as appropriate with the nutrient gradient. An 18-ml layer of 30 gr100 ml R2L of the appropriate nutrient level Žeither =0.01 or =1 R2 nutrient, depending on the nutrient gradient being prepared. was added by pouring and left to set at 248C for 30 min. A second aliquot of 18 ml 30 gr100 ml R2L Ž=0.1 R2 nutrient. was warmed slightly by holding at 248C for 10 min, poured on top of the initial layer and allowed to set. A final 18-ml aliquot of warmed 30 gr100 ml R2L of the appro-

143

priate nutrient level was then added and allowed to set. The three nutrient layers showed minimal mixing, and were clearly visible ŽFig. 2A.. Finally, the whole apparatus was placed at 58C for 1 h, to liquefy the gradient medium, and allow motile bacteria to move. The apparatus was then returned to 248C, and incubated for 48 h. The medium solidified, trapping cells in place, and colony development allowed assessment of movement. After initial data had established that the threelayer nutrient gradient system in the purpose-built plates was effective Ži.e. it was possible to form the gradient, and that motile heterotrophs would respond to this gradient., the approach was also tested in glass bottles in order to avoid the need for purposebuilt vessels and make the method more accessible; 80-ml CSW was pipetted into the bottom of sterile glass bottles. Bottles of 50-, 75- and 100-mm height and 25-mm diameter were tested. Three layers of 30 gr100 ml R2L of differing nutrient concentrations were prepared, with the nutrient gradient always running from low Žbottom layer. to high Žtop layer.. Each layer was set by incubation at 248C for 10 min, prior to adding the next layer. Again, the three layers remained distinct and separate ŽFig. 2B.. The system was then liquified and re-solidified as above. Control measures included gradient vessels inoculated with a suspension of a known motile, and a known nonmotile isolate, before incubation and sub-

Fig. 1. Diagram of the purpose-built culture vessels used in this study. Vessels were constructed from 5-mm clear perspex. ŽA. Main culture vessel; ŽB. 1-mm-thick butyl rubber gasket, cut to fit flush with internal edges of main culture vessel; ŽC. face plate. The three parts were held together using 3-mm diameter screw-threaded bolts, with finger-tightened nuts Žwing nuts..

144

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

Fig. 2. Photographs of nutrient gradient culture systems. ŽA. Photograph of purpose-built culture vessel showing the three layers of Lutrol F127, with the 1 = R2L layer at the top, before liquefaction. In this case, inoculation was at the bottom of the vessel and motile bacteria were removed from the top layer. ŽB. Photograph of nutrient gradient system prepared in a glass bottle. In this system, inoculation has to be at the bottom of the bottle and sampling is only possible from the top.

sequent sampling, to test for nonspecific movement through mixing and convection currents within the vessels. Both isolates were obtained from Taddiforde Brook after plating on R2A and testing random colonies for motility. 2.3.2. Preparation of horizontal nutrient gradient culture systems Horizontal nutrient gradients were prepared to confirm motility in Lutrol F127. This was achieved by culturing the motile and nonmotile isolates on

R2A. Individual, isolated colonies were suspended in 1-ml sterile water; 0.5 ml of these suspensions was added to 0.5-ml sterile glycerol Žto minimise movement during inoculation. and mixed by vortexing. Cylinders of R2 agar were obtained using alcoholsterilised cork borers. Six-well tissue culture plates ŽFalcon, Becton Dickinson, NJ. were used as culture vessels, with an R2A cylinder placed against an edge of each well, and the wells filled with 30% unamended Lutrol F127. Each of five wells contained an R2A cylinder of a different diameter, ranging from 4

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

145

to 15 mm. A control well in each plate contained a 15-mm 1.5% wrv water agar cylinder Ži.e. distilled water solidified with agar.. Each well was inoculated at the edge directly opposite the R2A cylinder with 20 ml of the glycerol suspension of each isolate. Plates were incubated at 58C for 1 h before incubation at 248C for 2 days. 2.4. Motility testing of isolates which responded to the nutrient gradient After incubation at 248C for 2 days to form colonies, the nutrient gradient vessels were sampled to test for motility of the isolates. Colonies from the purpose-built vertical gradient plates were sampled by removing the faceplate to expose the plastic sheet covering the medium. This was peeled back and individual colonies were removed using a flamed metal spike Žcooled by stabbing into an R2A plate.. Isolates were streaked onto R2A and incubated at 248C for 2 days and checked for purity. For both the purpose-built plates and the bottles, colonies were taken from the top layer, streaked onto R2A and tested for motility after 2 days, using the hanging drop method with bright field microscopy at 1000 = magnification. Colonies from further down the medium gradient were also tested from the purposebuilt plates. Isolated colonies from traditional spread plates, using 100-ml stream water on R2A, were tested for motility to establish background motility levels.

3. Results and discussion 3.1. Characteristics of Lutrol F127 as a gelling agent in culture media Inoculating plates of pure Lutrol F127 with either CSW or suspensions of laboratory grown bacteria demonstrated that Lutrol F127 did not support bacterial growth Žas judged by colony formation., unless supplemented with a nutrient source. A calibration curve showing gelling temperature with Lutrol F127 concentration was prepared ŽFig. 3.. For convenience, a concentration of 30-g Lutrol F127 in 100-ml water was used, as this was uniformly solid at room

Fig. 3. Calibration curve of gelling temperature against concentration of Lutrol F127.

temperature Ž248C. and liquid at 58C. This therefore avoided exposing environmental bacteria to higher temperatures whilst maintaining sufficient fluidity to allow bacterial movement. 3.2. Bacterial motility through Lutrol F127 as a response to a nutrient gradient It was demonstrated that bacterial motility was possible through a solution of 30 gr100 ml Lutrol F127 by producing horizontal R2 nutrient gradients in tissue culture plates. A halo of bacterial growth was only obtained around R2A cylinders of greater than 6-mm diameter, suggesting that a minimum volume of R2A was required to provide sufficient diffusable nutrient for detectable cell growth. When wells were inoculated with nonmotile bacterial suspensions, no such halos were obtained. In experiments with nonmotile bacteria, growth was only observed in wells containing the largest R2A cylinder Ž15-mm diameter., presumably due to the increased nutrient levels produced by diffusion. Nutrient gradients were produced using the purpose-built plates. Inoculation of the plates with CSW at the bottom Žbelow the 0.01 = R2L layer. with a nutrient gradient that increased vertically, resulted in a well-defined line of bacterial growth at the top of the culture medium. This indicated that motile bacteria had moved along the gradient of increasing nutrient concentration, and presumably also towards an increased oxygen concentration. When the nutrient

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

146

gradient was inverted, and the plates inoculated on the top Ž0.01 = R2L., a line of growth was obtained between the bottom and middle layers Ž1 = R2L and 0.1 = R2L, respectively.. This demonstrates that the motile bacteria again moved specifically towards the higher nutrient concentration, but in this case derived no oxygen advantage. Colonies were removed and subcultured from these positions and motility was confirmed. The proportion of motile bacteria isolated from these lines of growth is presented in Table 1. An increase in the proportion of motile bacteria isolated is clearly demonstrated. Vessels inoculated with nonmotile cells did not show any evidence of movement in response to the nutrient gradient Ži.e. no colonies were formed in the high nutrient layer.. Mixing CSW with nonmotile cells to give ) 99% nonmotile cells in the inoculum yielded only motile cells at the top of the nutrient gradient Ži.e. movement in specific response to nutrient concentrations.. Gradient formation in these plates was not completely reliable, as after liquefaction, mixing of layers sometimes occurred. To overcome this unreliability and to make the method more widely accessible Žby avoiding the need for purpose-built vessels., the use of nutrient gradients in glass bottles was also investigated ŽFig. 2B.. This approach limited accurate sampling to the top of the medium, and therefore the nutrient gradient was always increased vertically, with inoculation at the bottom. Again, an increased proportion of motile bacteria were isolated from the top layer ŽTable 1.. The vertical height of the gradient media appeared to play a role in successful isolation of motile bacteria. As the height increased, a larger proportion of motile bacteria were isolated ŽTable 1.. All isolates tested from vessels filled with

gradient media to a depth greater than 6 cm were motile, suggesting that a minimum distance for movement is required to successfully separate motile cells. The occurrence of nonmotile bacteria near the surface of nutrient gradient preparations below this 6-cm minimum distance can be explained by nonspecific cell movement, via physical disturbance from handling, convection currents and gradient preparation. Different bacteria will show different motility responses to different stimuli. The isolates obtained using the described methods are obviously responding in a positive fashion to the nutrient supplied in the R2B, at temperatures of around 58C. Using other nutrient sources and different concentrations of Lutrol F127 ŽFig. 3., it would be possible to alter gradient conditions and temperatures to target cells that are not motile under the described conditions. Additionally, although not tested here, responses to negative stimuli Ži.e. repellants. could be examined under similar conditions, by supplementing the Lutrol F127 with nutrient to allow cell division to occur after movement. Isolation of motile bacteria is of general interest for ecological studies, and for the targeted isolation of pathogens from heterogeneous cell suspensions. The method described above offers several advantages for such isolation. It is a simple, straightforward protocol to follow, yet is highly effective. The cost of the Lutrol F127 is comparable to that of agar. The ability to control the solidification of the medium at a low temperature offers the possibility of avoiding heat shock associated with use of agar in some techniques Že.g. pour plates.. It may ultimately prove to be possible to target the selectivity of the media, e.g. through the use of fluorogenic Lutrol F127 media to isolate motile b-D-glucuronidase-positive cells.

Table 1 Numbers of motile bacteria isolated before and after selection using different methods Isolate origin

Number of water Number of %Motile samples tested isolates tested

Stream water Purpose-built plates 12-ml glass bottles 18-ml glass bottles 27-ml glass bottles

4 4 2 1 1

76 30 53 50 44

42 100 81 98 100

Acknowledgements Rebecca Thomson was supported by a studentship from the Society for General Microbiology, UK. The authors wish to thank Clive Good for assistance with photography.

R. Thomson et al.r Journal of Microbiological Methods 46 (2001) 141–147

References Gardener, S., Jones, J.G., 1985. A new solidifying agent for culture media which liquefies on cooling. J. Gen. Microbiol. 130, 731–733. Gilbert, P., Jones, M.V., Allison, D.G., Heys, S., Maira, T., Wood, P., 1998. The use of poloxamer hydrogels for the assessment of biofilm susceptibility towards biocide treatments. J. Appl. Microbiol. 85, 985–990. Merino, S., Rubires, X., Aguilar, A., Tomas, J.M., 1997. The role of flagella and motility in the adherence and invasion to fish cell lines by Aeromonas hydrophila serogroup O:34 strains. FEMS Microbiol. Lett. 151, 213–217. Ojima, M., Tamagawa, H., Hayashi, N., Hanioka, T., Shizukuishi, S., 1998. Semi-automated measurement of motility of human subgingival microflora by image analysis. J. Clin. Periodontol. 25, 612–616.

147

Olson, L.D., 1996. Enhanced isolation of Serpulina hyodysenteriae by using sliced agar media. J. Clin. Microbiol. 34, 2937–2941. Peters, A.C., Thomas, L., Wimpenny, J.W.T., 1991. Effects of salt concentration of bacterial growth on plates with gradients of pH and temperature. FEMS Microbiol. Lett. 77, 309–314. Poppe, C., Duncan, C.L., 1996. Comparison of detection of Salmonella by the Tecraw Uniquee Salmonella test and the modified Rappaport Vassiliadis medium. Food Microbiol. 13, 75–81. Reasoner, D.J., Geldreich, E.E., 1985. A new medium for the enumeration and subculture of bacteria from potable water. Appl. Environ. Microbiol. 49, 1–7. Wirtanen, G., Salo, S., Allison, D.G., Mattila-Sandholm, T., Gilbert, P., 1998. Performance evaluation of disinfectant formulations using poloxamer-hydrogel biofilm-constructs. J. Appl. Microbiol. 85, 965–971.