- Email: [email protected]
CTC staining and counting of actively respiring bacteria in natural stone using confocal laser scanning microscopy
Journal of Microbiological Methods 52 (2003) 75 – 84 www.elsevier.com/locate/jmicmeth
CTC staining and counting of actively respiring bacteria in natural stone using confocal laser scanning microscopy S. Bartosch a,*,1, R. Mansch b,1, K. Kno¨tzsch c, E. Bock c a
Microbiology and Gut Biology Group, Ninewells Hospital and Medical School, University of Dundee, Dundee, DD1 9SY, UK b Bereich Humanmedizin, Universita¨t Go¨ttingen, D-37075, Germany c Institut fu¨r Allgemeine Botanik, Universita¨t Hamburg, D-22609, Hamburg, Germany Received 27 February 2002; received in revised form 26 June 2002; accepted 26 June 2002
Abstract A method was established for staining and counting of actively respiring bacteria in natural stone by using the tetrazolium salt 5-cyano-2,3-ditolyltetrazolium chloride (CTC) in combination with confocal laser scanning microscopy (CLSM). Applying 5 mM CTC for 2 h to pure cultures of representative stone-inhabiting microorganisms showed that chemoorganotrophic bacteria and fungi—in contrast to lithoautotrophic nitrifying bacteria—were able to reduce CTC to CTF, the red fluorescing formazan crystals of CTC. Optimal staining conditions for microorganisms in stone material were found to be 15 mM CTC applied for 24 h. The cells could be visualized on transparent and nontransparent mineral materials by means of CLSM. A semi-automated method was used to count the cells within the pore system of the stone. The percentage of CTC-stained bacteria was dependent on temperature and humidity of the material. At 28 jC and high humidity (maximum water holding capacity) in the laboratory, about 58% of the total bacterial microflora was active. On natural stone exposed for 9 years at an urban exposure site in Germany, 52 – 56% of the bacterial microflora was active at the east, west, and north side of the specimen, while only 18% cells were active at the south side. This is consistent with microclimatic differences on the south side which was more exposed to sunshine thus causing UV and water stress as well as higher temperatures on a microscale level. In combination with CLSM, staining by CTC can be used as a fast method for monitoring the metabolic activity of chemoorganotrophic bacteria in monuments, buildings of historic interest or any art objects of natural stone. Due to the small size of samples required, the damage to these objects and buildings can be minimized. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Acridine orange; Cell counts; Chemoorgaotrophic bacteria; CLSM; CTC; Natural stone; Nitrifying bacteria
Abbreviations: AO, acridine orange; CLSM, confocal laser scanning microscopy; CTC, 5-cyano-2,3-ditolyltetrazolium chloride; CTF, formazan crystals of CTC; DAPI, 4,6-diamidino-2-phenylindol; INT, 2-( p-iodophenyl)-3-( p-nitrophenyl)-5-phenyltetrazolium chloride; LC, laboratory-colonized specimen; LC+NE, laboratory-colonized and naturally exposed specimen; MPN, most probable number; TTC, 2,3,5triphenyltetrazolium chloride. * Corresponding author. Tel.: +44-1382-496341; fax: +44-1382-633952. E-mail address: [email protected] (S. Bartosch). 1 The work presented has been done at the University of Hamburg. The addresses given are the current addresses of the authors. 0167-7012/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 7 0 1 2 ( 0 2 ) 0 0 1 3 3 - 1
76
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
1. Introduction The contribution of microorganisms to the deterioration of natural stone, mortar and concrete has been widely investigated during the last 10 – 15 years (Griffin et al., 1991; Lyalikova and Petrushkova, 1991; Ortega-Calvo et al., 1991; Palmer et al., 1991; De la Torre et al., 1993; May et al., 1993; Hirsch et al., 1995a; Warscheid and Krumbein, 1996; Mansch and Bock, 1998; Warscheid and Braams, 2000). Chemoorganotrophic bacteria, fungi and chemolithoautotrophic nitrifying bacteria represent the epi- and endolithic microflora most relevant to the biodeterioration of natural building stone (Warscheid and Krumbein, 1996; Mansch and Bock, 1998). To evaluate the danger to historical buildings, monuments or statues from biodeterioration, it is essential to quantify these microorganisms. Enumeration is usually done by plate count and most probable number (MPN) techniques, which are based on the cultivation of microorganisms on selective media. These techniques need a relatively high amount of sample material, and are quite timeconsuming requiring 1 –6 weeks of incubation. Due to suboptimal culture conditions and methodological limitations, the cell numbers are usually underestimated. An overview of destructive and nondestructive methods used to study the stone-inhabiting microflora in rocks and building stones has been given by Hirsch et al. (1995b). In comparison to cultivation techniques, microscopic cell counts can be much faster for the evaluation of microbial colonization and usually higher cell numbers are found (Richaume et al., 1993). Confocal laser scanning microscopy (CLSM) offers the opportunity to visualize microorganisms in situ within the pore system of mineral materials (Quader and Bock, 1995; Bartosch et al., 2002). Fluorescence signals can be measured by the digital image analysis system which allows semi-automatic counting of microorganisms in stone material (Bartosch et al., 1996). However, many fluorescence dyes including DAPI which is widely applied to study bacteria in natural samples proved to be unsuitable, as these dyes bind to the stone material and cause strong background fluorescence which hinders visualization of the microorganisms (unpublished). Only staining by means of acridine orange (AO) has been shown to be suitable to visualize microorganisms on and in natural stone
(Quader and Bock, 1995). After dye application, the green fluorescing microorganisms can be easily distinguished from the red fluorescing mineral components (e.g. clay particles). However, with AOstaining, it is not possible to distinguish between active and inactive microorganisms, as they are all stained green, whereas after autoclaving the stone material, dead cells show red fluorescence. Thus, dead cells can hardly be distinguished from red fluorescing stone matrix. For assessment and monitoring of the risk of biodeterioration of a historical building, it may be necessary to know the in situ activity of microorganisms. Tetrazolium salts are commonly used to analyze the activity of natural populations. They act as artificial electron acceptors within functional electron transport systems (e.g. respiratory) or for certain active dehydrogenases. So far, little work has been done on activity analysis of stone-inhabiting microorganisms. The tetrazolium salts 2,3,5-triphenyltetrazolium chloride (TTC) (Warscheid et al., 1990) and 2-( p-iodophenyl)-3-( p-nitrophenyl)-5-phenyltetrazolium chloride (INT) (Taylor and May, 1995, 2000) have been used to investigate microbial activity in stone. These approaches were limited by the fact that the reduction products have to be extracted for quantitative study and this usually underestimates the cell numbers. In the study presented here, the tetrazolium salt 5cyano-2,3-ditolyltetrazolium chloride (CTC) has been used for the first time to visualize and quantify in situ the actively respiring microorganisms in natural stone. In previous studies, CTC had been applied successfully to quantify active bacteria in groundwater, fresh water and seawater (Rodriguez et al., 1992; Sherr et al., 1999), drinking water (Schaule et al., 1993; Kalmbach et al., 1997) activated sludge (Rodriguez et al., 1992; Griebe et al., 1995), estuarine sediments (Proctor and Souza, 2001) and soil (Winding et al., 1994; Yu et al., 1995). In contrast to INT and TTC, the formazan crystals formed by CTC reduction (CTF) show red fluorescence and CTC is therefore suitable for the visualization of microorganisms in stone using CLSM. To establish a method for selective counting of active microorganisms, representative pure cultures of stone-inhabiting microorganisms were checked for their ability to reduce CTC to CTF. The staining
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
conditions for microorganisms in natural stone were optimized and the influence of temperature and humidity on the microbial activity was studied. CTC staining was applied to determine the percentage of active bacteria in natural stone at an urban exposure site in Germany.
2. Materials and methods 2.1. Bacterial strains and culture conditions Pure cultures of Micrococcus varians ATTC 15306 and Paracoccus denitrificans ATTC 19367 were obtained from the German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany. All other strains used in this study had been isolated from building stone. The chemolithoautotrophic bacteria Nitrosovibrio spec. K7.1 and Nitrobacter vulgaris K55, as well as the facultatively methylotrophic bacterium Methylobacterium fuyisawaense L5, were isolated from Cologne Cathedral, Germany. The actinomycete Geodermatophilus spec. BC 532 was isolated from natural stone of the church Santa Maria Scala Quarry in Noto, Italy and kindly supplied by Dr. Clara Urzı´, University of Messina, Italy. The yeast Sporobolomyces roseus Kluyver and Niel, the filamentous fungi Phoma leveillei Boerema and Bollen, and Trichoderma citrinoviride Bissett were isolated from stone specimens at an exposure site in Duisburg, Germany. All strains are stored in the culture collection of the Institut fu¨r Allgemeine Botanik at the University of Hamburg, Germany. M. varians ATTC 15306, P. denitrificans ATTC 19367, and Geodermatophilus spec. BC 532 were cultivated on Standard I nutrient broth (Merck No. 107882). Nitrosovibrio spec. K7.1 was cultivated in a mineral medium with 10 mM NH4Cl (Koops et al., 1991). N. vulgaris K55 was grown lithoautotrophically in a mineral medium with 30 mM NaNO2 (Bock et al., 1990). M. fuyisawaense L5 was grown in a basal salt solution of pH 8.4 with 1.5% methanol according to Green et al. (1988). For cultivation of S. roseus Kluyver and Niel, P. leveillei Boerema and Bollen, and T. citrinoviride Bissett, a Sabouraud-Maltose agar (Merck No. 5439) was used. For CTC analysis, these fungi were transferred and grown in Malt extract broth (Merck No. 105397).
77
2.2. Stone samples and experimental conditions All investigations were done using Ihrlersteiner Green Sandstone. This type of sandstone is a dolomitic material, which supports the growth of chemoorganotrophic bacteria and lithoautotrophic nitrifying bacteria (Mansch and Bock, 1998). Stone samples 5 mm thick were taken from the stone specimens for investigation by using a diamond-coated cutting disc (Ø 20 0.6 mm thick) attached to a Proxxon hobbydriller. Stone specimens of size 1.5 1.5 1.5 cm were initially inoculated in the laboratory with pure cultures of ammonia and nitrite oxidizers and kept for several years in a plastic box on fine grain sand at maximum water holding capacity in an atmosphere of about 98% relative humidity and at 28 jC. Due to the nonsterile conditions, a complex microflora developed on these specimens. These specimens were used to determine the optimal CTC concentration and incubation time and are referred to as laboratory-colonized (LC) specimens. The influence of drying on the number of active bacteria was determined on the above-described LC specimens. During drying, the specimens were stored at 28 jC and 40% relative humidity and were analyzed after 0, 1, 2, and 5 days. After 5 days, the samples were remoistened with sterile water and analyzed after 6 h. To analyze the influence of temperature on the number of active bacteria, LC specimens of size 1.5 5 5 cm were used. These specimens had been additionally exposed for 2 years to outside conditions in the botanical garden of the Institut fu¨r Allgemeine Botanik at the University of Hamburg, Germany and are referred to as laboratory-colonized and naturally exposed (LC + NE) specimens. During this exposure algae had grown on the stone surface. Prior to CTC staining, the samples were incubated at different temperatures (5, 10, 28, and 35 jC) for 4 weeks at high stone moisture (maximum water holding capacity) and in an atmosphere of about 98% relative humidity. CTC staining on naturally colonized stone was carried out on samples taken from an urban exposure site in Duisburg, Germany on three different occasions in summer 1997. The site has been previously described by Mansch and Bock (1998). The specimen
78
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
sampled for this study was 20 30 45 cm and had been exposed for 9 years. Six samples were collected from vertical surfaces exposed to the east, west, north, and south side of the stone specimen, respectively, on 5th August 1997. At that time, it had not rained for 4 days and the daytime temperature ranged between 20 and 25 jC. During transport, the samples were kept in small containers to protect them from breaking and were sealed to avoid drying out of the natural moisture of the stone material. Transport took approximately 6 h and samples were stained immediately on arrival in the laboratory. 2.3. Staining with CTC Pure liquid culture samples (1 ml) were taken aseptically at different stages of growth (lag, exponential, and stationary phase). The growth phases were checked by microscopic cell counts using a Helber chamber for M. varians, P. denitrificans, M. fuyisawaense, S. roseus, N. vulgaris, and Nitrosovibrio spec. Cell protein was determined colorimetrically by using the method of Bradford (1976), as modified by Spector (1978) for Geodermatophilus spec., P. leveillei, and T. citrinoviride. About 100 Al of a 50 mM CTC solution (Polyscience Europe) was added to the samples and incubated for 2 h at room temperature in the dark. The cells were then washed with sterile 0.9% NaCl and suspended in 500 Al of the 0.9% NaCl solution, and 500 Al of a 4% formaldehyde solution was added to stop CTC reduction. After fixation for 10 min on ice, cells were washed twice with 0.9% NaCl. Counterstaining of total cells was done with 4,6,diamidino-2-phenylindol (DAPI) (10 Ag ml 1) for 5 min. Optimal CTC concentration and incubation time for staining of microorganisms in stone were tested. The following procedure achieved the best result. Freshly cut stone samples with a size of 5 5 5 mm were stained with 100 Al of a 15 mM CTC solution following a 24-h incubation period at room temperature in the dark. During incubation, the samples were kept in Eppendorf-tubes so that they were covered by the CTC solution. To stop CTC reduction, the stone samples were dried in air and stored at 20 jC. Control samples were treated with 2% formaldehyde or 0.1% sodium azide prior to CTC application. Staining of total cells was performed on parallel
samples with 200 Ag l 1 AO dissolved in distilled water containing 1% dimethylsulfoxid (Merck) for 1 min. 2.4. Microscopic determination of cell numbers Stained samples of pure cultures were filtered onto a polycarbonate filter with a pore size of 0.2 Am (Costar, Bodenheim) according to Hobbie et al. (1976). Cell numbers were determined with a Leica DM IRBE inverse microscope using a PL Fluotar objective (40 /1.0/oil) and an ocular L Plan (10 ). A Leica filter set N 2.1 (BP 515– 560 exc./RKP 580/ LP 590 em.) was used for CTF detection. DAPI was visualized with a Leica filter set A (BP 340– 380 exc./ RKP 400/LP 425 em.). Cell counts were carried out within a square of an area of 6.25 105 Am2 inside the ocular. At least 400 cells on 20 randomly chosen areas on a filter where counted and the mean value and standard deviation was calculated. Microscopic analysis of stone samples was done on aluminum slides prepared with a drill hole in the center to which a cover glass was fixed. The stained samples were placed in a drop of water on the cover glass and examined with an inverse microscope. Cell numbers were determined on transverse sections cut at 90j to the stone surface. On six parallel stone samples, five cell counts were done on different areas at various depths in the ranges 0 –0.5, 0.5– 2.0, and 2.0– 5.0 mm to achieve the mean cell number for a layer of 5-mm overall depth. The stone samples were investigated with a CLSM (Leica model TCS 4D) using a PL Fluotar objective (40x/1.0/oil). CTF was excited at 568 nm with an argon krypton laser and the emitted red fluorescence was detected with a long pass filter at 590 nm. AOstained microorganisms were excited at 488 nm and the emitted green fluorescence detected with a band pass filter at 520 – 560 nm. Simultaneous staining with CTC and AO is not possible as AO also shows red fluorescence when it is excited with the 568-nm wavelength and so CTC and AO-staining was done on different subsamples. Laser power, pinhole, and photo multiplier were adjusted to bright fluorescence of the microorganisms with the background showing no fluorescence. A series of 25 optical sections at a distance of 1 Am apart was recorded for each cell count. The total fluorescence intensity of the stone
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
sample was measured by means of the image analysis program (SCANware 5.1, Leica) within an area of 250 250 Am covering a depth of 25 Am. The mean fluorescence intensity of three single bacterial cells was measured and the cell number of the scanned volume unit was calculated as follows: cell number total fluorescence intensity within measured space ¼ mean bacterial fluorescence intensity of a single cell
Using these data, the cell numbers per gram of stone material were extrapolated.
3. Results Staining of pure cultures was achieved with 5 mM CTC, incubated for 2 h at room temperature. As shown in Table 1, chemoorganotrophic bacteria and fungi were able to reduce CTC to red fluorescing CTF, while lithoautotrophic nitrifying bacteria could not be stained. In the case of M. varians, P. denitrificans, M. fuyisawaense, and S. roseus, the percentage of CTF-positive organisms was correlated to the growth phase. During the lag phase, 15 –30% of total cells were stained. The percentages of CTF-positive cells increased during the exponential growth phase to 65 –79% of total cells. During the stationary phase, the number of CTF-positive cells decreased to 15– 28% of total cells. It was not possible to quantify
79
CTF-positive cells of Geodermatophilus spec., P. leveillei, and T. citrinoviride with the microscope because they are filamentous organisms. Therefore, the growth phase was determined by measuring calorimetrically the increase in cell protein. Microscopic observations indicated that CTF crystals accumulated in the cells at approximately the same rate at each stage of their growth phase. In contrast to pure cultures, optimal staining of bacteria on natural stone was achieved at a CTC concentration of 15 mM and after an incubation of 24 h. Single cells stained by CTC could be detected on LC specimens (Fig. 1a). Bacterial microcolonies were found on naturally colonized stone derived from the exposure site in Duisburg (Fig. 1b). Control samples treated with formaldehyde or sodium azide showed no fluorescence. When LC samples were dried, the number of CTFpositive cells decreased (Fig. 2), while the total cell numbers determined with AO remained constant (6.5 109 cells g 1 stone). Under wet conditions, 60% of total cells were stained with CTC. About 39% of total cells were still active after 1 day of drying, and after 2 days of drying, the percentage of CTF-positive cells decreased to 20%. This level did not change during three more days of drying. Remoistening of the samples after 5 days of drying resulted in an increase of active cells to 48% after 6 h. The influence of temperature on the number of active cells on LC + NE specimen is shown in Fig. 3.
Table 1 CTC reduction of representative stone-inhabiting (a) chemoorganotrophic bacteria, (b) fungi, and (c) nitrifying bacteria Organisms
(a)
(b)
(c)
Micrococcus varians Paracoccus denitrificans Methylobacterium fuyisawaense Geodermatophilus spec. Sporobolomyces roseus Phoma leveillei Trichoderma citrinoviride Nitrosovibrio spec. Nitrobacter vulgaris
CTC reduction
CTC reduction of sterilized cells
Percentage of CTF positive cells during the exponential phase
+ + + + + + +
66 – 78% 68 – 79% 68 – 79% n.a. 65 – 68% n.a. n.a. n.d. n.d.
Microscopic cell counts of CTF positive cells of Geodermadophilus spec., Phoma leveillei, and Tichoderma citronoviride could not be determined because they are filamentous organisms. n.a. = not analyzed. n.d. = not detected.
80
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
Fig. 2. Influence of drying (28 jC, 40% relative humidity) on the percentage of total (AO) and active (CTC) cells counted on laboratory-colonized specimens of Ihrlersteiner Green Sandstone incubated at high stone moisture (5R) (maximum water holding capacity). After remoistening content the activity of the cells recovered.
A maximum of 60% active cells was found at 28 jC, whereas at 5, 10, and 35 jC, the percentages of active cells were considerably lower than at 28 jC with values between 25% and 32% of CTC-stained cells. AO staining revealed that the total cell numbers stayed at a constant level of about 9.1 107 cells g 1 stone for all temperatures studied. On naturally colonized Ihrlersteiner Green Sandstone from the exposure site in Duisburg, 18 – 56% of the total cells were stained by CTC. The measured CTC reduction activity could be attributed mainly to bacteria because fungi were rarely detected during microscopic investigations. On the north-, east-, and
Fig. 1. Confocal laser scanning micrographs of CTC-stained bacteria on natural stone. (a) Single red fluorescing bacteria (B) distributed within the pore system of a laboratory-colonized specimen. The stone (S) material showed slight green autofluorescence. (b) Red fluorescing bacterial microcolony (M) visualized on natural stone colonized under natural conditions. Bars = 5 Am. CTC-stained cells and stone autofluorescence were visualized with a CLSM (microscope: Leica model TCS 4D); excitation was provided by an argon krypton laser.
Fig. 3. Influence of temperature on the percentage of active (CTC) and total (AO) cells on laboratory-colonized stone specimen, which had been exposed additionally to outside conditions. The specimens of Ihrlersteiner Green Sandstone had been stored for 4 weeks at different temperatures before staining.
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
Fig. 4. Active and total cell numbers at different sides of a specimen from Ihrlersteiner Green Sandstone exposed for 9 years at an urban exposure site in Duisburg, Germany. Sampling took place on 5th August 1997 after 4 days without rain and a daytime temperature ranging between 20 and 25 jC.
west-oriented sides, a nearly constant level of 52– 56% of total bacterial cells (average 6.5 107 cells g 1 stone) were CTF-positive (Fig. 4), whereas on the south side, only 18% of 6.7 106 cells g 1 stone were found to be active.
4. Discussion Pure culture experiments with selected strains belonging to the stone-inhabiting microflora showed that chemoorganotrophic bacteria and fungi were able to reduce the redox dye CTC to CTF. The optimal staining conditions were found to be 5 mM CTC incubated for 2 h (data not shown) which corresponds to previous findings (Griebe et al., 1995; Choi et al., 1999). For the chemoorganotrophic bacteria and S. roseus, the amount of cells reducing CTC was dependent on the growth phase, whereby cells from the exponential phase showed the highest activity similar to previous findings (Sherr et al., 1999). In contrast, the lithoautotrophic nitrifying bacteria could not be stained by CTC. CTC is reduced in the presence of functional electron transport (i.e. respiratory) systems or certain active dehydrogenases. Although ammonia and nitrite oxidizers possess high amount of electron transport proteins (Bock et al., 1991), we hypothesize that the redox potential of the electron carriers was too low to reduce a detectable amount of CTC under the given staining conditions. Increased incubation time
81
up to 24 h and CTC concentration up to 15 mM (data not shown) did not alter the results. However, under these conditions, toxic effects have to be taken into account as previously reported for liquid cultures and seawater samples (Griebe et al., 1995; Choi et al., 1999). We demonstrated that microorganisms in natural stone could be stained by CTC. Control samples treated with biocides (formaldehyde, azide) clearly showed that chemical CTC reduction can be excluded. Indeed, bacterial CTC reduction correlated with temperature which has already been shown to influence the activity of marina bacteria (Choi et al., 1999) and water stress. A similar correlation between bacterial activity, water stress, and temperature has also been observed for soil samples (Winding et al., 1994). These findings therefore provide clear evidence that the CTC reduction in stone was induced by microbial activity and the findings were also confirmed by the analysis of the specimens from the exposure site in Duisburg. Under natural conditions, the lowest numbers of active bacteria were found at the south side which is more exposed to sunshine. Thus, these microorganisms were more influenced by UV light and the microclimate in the stone was very likely characterized by increased water stress and intermittent high temperatures which can have a negative effect on microbial activity above 28 jC as demonstrated on the laboratory-colonized specimens (Fig. 3). On natural stone a much higher CTC concentration and longer incubation time was necessary for the staining of bacteria than in pure cultures. A CTC concentration of 15 mM obviously did not have toxic effects on the stone-inhabiting microorganisms, although toxic effects have been observed for liquid cultures at concentrations above 10 mM (Griebe et al., 1995). For seawater samples, a slight decrease of CTF positive cells was observed when 15 mM CTC was used compared to 5 mM CTC, but CTC reduction still took place (Choi et al., 1999). Previous investigations on soil samples showed that longer incubation times were required for successful staining compared to the conditions used for analysis of liquid cultures (Yu et al., 1995). Two factors may be responsible for this finding on stone material: (i) a certain percentage of CTC might be adsorbed at the pore surface of the mineral matrix; (ii) extracellular polymeric substances, which usually surround bacteria growing attached
82
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
to mineral surfaces, might have hindered the penetration of CTC into the cells. CLSM cell counts of microorganisms in stone were attributed mainly to the chemoorganotrophic bacterial microflora, whereas nitrifying bacteria, the main representatives of the lithoautotrophic stone-inhabiting microflora, and fungi hardly played a role during microscopic quantification. In fact, cell number determination of microorganisms in stone using CLSM is a fairly insensitive method with a detection limit of 105 cells g 1 stone (Bartosch et al., 1996) compared to plate or MPN counts, which have a detection limit between 101 and 102 CFU g 1 stone depending on the methodological approach. While chemoorganotrophic bacteria are frequently found in natural building stone using cultivation techniques at cell numbers above 105 CFU g 1, fungi and nitrifying bacteria are detectable at lower cell numbers by at least one or three orders of magnitude respectively (Mansch and Bock, 1998). In over 500 samples from historical buildings containing nitrifying bacteria, the average cell number was found to be 3.8 101 cells g 1 stone for ammonia-oxidizers and 5.3 101 cells g 1 for nitrite oxidizers (Mansch and Bock, 1998) which is far below the microscopic detection limit. In addition, fungi, which can be distinguished morphologically from bacteria, were very rarely observed during the microscopic investigations in this study. However, the ability of some stone-inhabiting fungi to form thick, melaninized cell walls (Wollenzien et al., 1995) may prevent the penetration of dyes and should also be taken into consideration. In the case of heavy growth of filamentous fungi, the semi-automated quantification of active bacteria by CLSM would certainly be obstructed because it is impossible to differentiate between CTF crystals accumulated in fungi or those accumulated in bacteria. Algae were found on the light-exposed surfaces of LC + NE specimens and the naturally colonized specimen from Duisburg. On vertical facades of historical buildings, algae are usually restricted to those areas which are in direct contact with running water or affected by rising damp. Algae and cyanobacteria can be easily detected without the application of fluorescence dyes as their chlorophyll shows red autofluorescence. However, this fluorescence spectrum is similar to that of CTF, a problem which concerns many commercially available fluorescence dyes (Lie-
sle et al., 1999). For investigations of complex biofilms containing algae and cyanobacteria, the autofluorescing cells could be discriminated from CTC-stained cells by separating the red CTC fluorescence from the far red fluorescence originating from the chlorophyll containing organisms (Neu, 2000). However, for biodeterioration studies, it is necessary to analyze the microorganisms growing within the depth of the building material. For this reason, bacteria were counted on transverse sections from 0- to 5-mm depth because previous studies using cultivation techniques revealed that bacteria and fungi are found mainly within this depth (Mansch and Bock, 1998). On the analyzed transversal sections, algae were not detected and bacterial cell counts could be done unhindered using the red spectrum of the fluorescence light. In conclusion, CTC staining in combination with CLSM quantification was proven to be a suitable method to determine the in situ activity status of chemoorganotrophic bacteria in natural stone. A selective method to determine the state of activity of nitrifying bacteria in natural building stone is well established and has been widely applied (Mansch and Bock, 1996, 1998). This is a different approach whereby the end products of ammonia and nitrite oxidation, i.e. nitrite and nitrate, are measured in suspensions of ground stone material using high performance liquid chromatography. However, due to the variety of substrates, a comparable technique has not been described for chemoorganotrophic bacteria. CTC staining in combination with CLSM imaging is a quick and reliable method for the evaluation of the current microbial in situ activity on historical buildings. Only small samples are required and this is of great importance in minimizing the damage to such valuable objects. Therefore, this method is very promising in being helpful in assessing the risk of biodeterioration of historical buildings or, for example, to monitor the efficacy of preventive measures such as treatment with biocides.
Acknowledgements This study was supported by the German Ministry of Research Education and Technology (BMBF) through funding of the projects Bau 5016E and Bau 7016E.
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
References Bartosch, S., Quader, H., Bock, E., 1996. Confocal laser scanning microscopy: a new method for detecting microorganisms in natural stone. DECHEMA Monogr. 133, 37 – 43. Bartosch, S., Mansch, R., Pinck, C., Bock, E., 2002. Visualisation and quantification of microorganisms on mortar on natural stone by CLSM. In: Gala´n, E., Zezza, F. (Eds.), Protection and Conservation of the Cultural Heritage of the Mediterranean Cities. A.A. Balkema Publishers, Lisse, pp. 275 – 279. Bock, E., Koops, H.-P., Mo¨ller, U.C., Rudert, M., 1990. A new facultatively nitrite oxidizing bacterium, Nitrobacter vulgaris sp. nov. Arch. Microbiol. 153, 105 – 110. Bock, E., Koops, H.-P., Harms, H., Ahlers, B., 1991. The biochemistry of nitrifying bacteria. In: Shively, J.M., Barton, L.L. (Eds.), Variations in Autotrophic Life. Academic Press, London, pp. 171 – 200. Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle protein-dye binding. Anal. Biochem. 72, 249 – 254. Choi, J.-W., Sherr, B.F., Sherr, E.B., 1999. Dead or alive? A large fraction of ETS-inactive marine bacterioplancton cells, as assessed by reduction of CTC, can become ETS-active with incubation and substrate addition. Aquat. Microb. Ecol. 18, 105 – 115. De la Torre, M.A., Go´mez-Alarco´n, G., Vizcaino, C., Garcia, T., 1993. Biochemical mechanisms of stone alteration carried out by filamentous fungi living in monuments. Biogeochemistry 19, 129 – 147. Green, P.N., Bousfield, I.J., Hood, D., 1988. Three new Methylobacterium species: M. rhodesianum sp. nov., M. zatmani sp. nov., and M. fuyisawanease. Int. J. Syst. Bacteriol. 38, 124 – 127. Griebe, T., Schaule, G., Secker, J., Flemming, H.-C., 1995. Bestimmung der stoffwechselaktiven Bakterien im belebten Schlamm. Wasser Abwasser 136, 608 – 615. Griffin, P.S., Indictor, N., Koestler, R.J., 1991. The biodeterioration of stone: a review of deterioration mechanisms, conservation case histories, and treatment. Int. Biodeterior. 28, 187 – 207. Hirsch, P., Eckhardt, F.E.W., Palmer, R.J., 1995a. Fungi active in weathering of rock and stone monuments. Can. J. Bot. 73, 1384 – 1390. Hirsch, P., Eckhardt, F.E.W., Palmer, R.J., 1995b. Methods for the study of rock-inhabiting microorganisms—a mini review. J. Microbiol. Methods 23, 143 – 167. Hobbie, J.E., Daley, R.J., Jasper, S., 1976. Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33, 1225 – 1228. Kalmbach, S., Manz, W., Szewzyk, U., 1997. Dynamics of biofilm formation in drinking water: phylogenetic affiliation and metabolic potential of single cells assessed by formazan and in situ hybridization. FEMS Microbiol. Ecol. 22, 265 – 279. Koops, H.-P., Bo¨ttcher, B., Mo¨ller, U.C., Pommerening-Ro¨ser, A., Stehr, G., 1991. Classification of eight new species of ammonia oxidizing bacteria: Nitrosomonas communis sp. nov., Nm ureae sp. nov., Nm aestuarii sp. nov., Nm marina sp. nov., Nm nitrosa sp. nov., Nm eutropha sp. nov., Nm oligotropha sp. nov. and Nm halophila sp. nov. J. Gen. Microbiol. 137, 1689 – 1699.
83
Liesle, J.T., Stewart, P.S., McFeters, G.A., 1999. Fluorescent probes applied to physiological characterization of bacterial biofilms. In: Doyle, R.J. (Ed.), Biofilms: Methods in Enzymology. Academic Press, New York, pp. 166 – 178. Lyalikova, N.N., Petrushkova, Y.P., 1991. Role of microorganisms in the weathering of minerals in building stone of historical buildings. Geomicrobiol. J. 9, 91 – 101. Mansch, R., Bock, E., 1996. Simulation of microbial attack on natural and artificial stone. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Microbially Influenced Corrosion of Materials. Springer-Verlag, Berlin, pp. 167 – 186. Mansch, R., Bock, E., 1998. Biodeterioration of natural stone with special reference to nitrifying bacteria. Biodegradation 9, 47 – 64. May, E., Lewis, F.J., Pereira, S., Taylor, S., Seaward, M.R.D., Allsopp, D., 1993. Microbial deterioration of building stone—a review. Biodeterior. Abstr. 7, 109 – 123. Neu, T.R., 2000. In situ cell and glycoconjugate distribution in river snow studied by confocal laser scanning microscopy. Aquat. Microb. Ecol. 21, 85 – 95. Ortega-Calvo, J.J., Hernandez-Marine, M., Saiz-Jimenez, C., 1991. Biodeterioration of building materials by cyanobacteria and algae. Int. Biodeterior. 28, 165 – 185. Palmer Jr., R.J., Siebert, J., Hirsch, P., 1991. Biomass and organic acids in sandstone of a weathering building: production by bacterial and fungal isolates. Microb. Ecol. 21, 253 – 266. Proctor, L.M., Souza, A.C., 2001. Method for enumeration of 5cyano-2,3-ditolyltetrazolium chloride (CTC)-active cells and cell-specific CTC activity of benthic bacteria in riverine, estuarine and coastal sediments. J. Microbiol. Methods 43, 213 – 222. Quader, H., Bock, E., 1995. Confocal-laser-scanning microscopy: a new method to study microorganisms colonizing natural stone. Int. J. Restor. Build. 1, 295 – 304. Richaume, A., Steinberg, C., Jocteur-Monrozier, L., Faurie, G., 1993. Differences between direct and indirect enumeration of soil bacteria: the influence of soil structure and cell location. Soil Biol. Biochem. 25, 641 – 643. Rodriguez, G.G., Phipps, D., Ishiguro, K., Ridgway, H.F., 1992. Use of a fluorescent redox dye for direct visualisation of actively respiring bacteria. Appl. Environ. Microbiol. 58, 1801 – 1808. Schaule, G., Flemming, H.-C., Ridgway, H.F., 1993. Use of 5-cyano-2,3-ditolyltetrazolium chloride for quantifying planctonic and sessile respiring bacteria in drinking water. Appl. Environ. Microbiol. 59, 3850 – 3857. Sherr, B.F., Del Giorgio, P., Sherr, E.B., 1999. Estimating abundance and single-cell characteristics of respiring bacteria via the redox dye CTC. Aquat. Microb. Ecol. 18, 117 – 131. Spector, T., 1978. Refinement of Comassie-blue method of protein quantification. Anal. Biochem. 86, 142 – 146. Taylor, S., May, E., 1995. A comparison of methods for the measurement of microbial activity on stone. Stud. Conserv. 40, 163 – 170. Taylor, S., May, E., 2000. Investigations of the localisation of bacterial activity on sandstone from ancient monuments. Int. Biodeterior. Biodegrad. 46, 327 – 333. Warscheid, T., Braams, J., 2000. Biodeterioration of stone: a review. Int. Biodeterior. Biodegrad. 46, 343 – 368. Warscheid, T., Krumbein, W.E., 1996. General aspects and selected cases. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Micro-
84
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
bially Influenced Corrosion of Materials. Springer-Verlag, Berlin, pp. 273 – 295. Warscheid, T., Petersen, K., Krumbein, W.E., 1990. A rapid method to demonstrate and evaluate microbial activity on decaying sandstone. Stud. Conserv. 35, 137 – 147. Winding, A., Binnerup, S.J., Sørensen, J., 1994. Viability of indigenous soil bacteria assayed by respiratory activity and growth. Appl. Environ. Microbiol. 60, 2869 – 2875.
Wollenzien, U., De Hoog, G.S., Krumbein, W.E., Urzi, C., 1995. On the isolation of microcolonial fungi occurring on and in marble and other calcareous rocks. Sci. Total Environ. 167, 287 – 294. Yu, W., Dodds, W.K., Banks, M.K., Skalsky, J., Strauss, E.A., 1995. Optimal staining and sample storage time for direct microscopic enumeration of total and active bacteria in soil with two fluorescence dyes. Appl. Environ. Microbiol. 61, 3367 – 3372.
CTC staining and counting of actively respiring bacteria in natural stone using confocal laser scanning microscopy S. Bartosch a,*,1, R. Mansch b,1, K. Kno¨tzsch c, E. Bock c a
Microbiology and Gut Biology Group, Ninewells Hospital and Medical School, University of Dundee, Dundee, DD1 9SY, UK b Bereich Humanmedizin, Universita¨t Go¨ttingen, D-37075, Germany c Institut fu¨r Allgemeine Botanik, Universita¨t Hamburg, D-22609, Hamburg, Germany Received 27 February 2002; received in revised form 26 June 2002; accepted 26 June 2002
Abstract A method was established for staining and counting of actively respiring bacteria in natural stone by using the tetrazolium salt 5-cyano-2,3-ditolyltetrazolium chloride (CTC) in combination with confocal laser scanning microscopy (CLSM). Applying 5 mM CTC for 2 h to pure cultures of representative stone-inhabiting microorganisms showed that chemoorganotrophic bacteria and fungi—in contrast to lithoautotrophic nitrifying bacteria—were able to reduce CTC to CTF, the red fluorescing formazan crystals of CTC. Optimal staining conditions for microorganisms in stone material were found to be 15 mM CTC applied for 24 h. The cells could be visualized on transparent and nontransparent mineral materials by means of CLSM. A semi-automated method was used to count the cells within the pore system of the stone. The percentage of CTC-stained bacteria was dependent on temperature and humidity of the material. At 28 jC and high humidity (maximum water holding capacity) in the laboratory, about 58% of the total bacterial microflora was active. On natural stone exposed for 9 years at an urban exposure site in Germany, 52 – 56% of the bacterial microflora was active at the east, west, and north side of the specimen, while only 18% cells were active at the south side. This is consistent with microclimatic differences on the south side which was more exposed to sunshine thus causing UV and water stress as well as higher temperatures on a microscale level. In combination with CLSM, staining by CTC can be used as a fast method for monitoring the metabolic activity of chemoorganotrophic bacteria in monuments, buildings of historic interest or any art objects of natural stone. Due to the small size of samples required, the damage to these objects and buildings can be minimized. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Acridine orange; Cell counts; Chemoorgaotrophic bacteria; CLSM; CTC; Natural stone; Nitrifying bacteria
Abbreviations: AO, acridine orange; CLSM, confocal laser scanning microscopy; CTC, 5-cyano-2,3-ditolyltetrazolium chloride; CTF, formazan crystals of CTC; DAPI, 4,6-diamidino-2-phenylindol; INT, 2-( p-iodophenyl)-3-( p-nitrophenyl)-5-phenyltetrazolium chloride; LC, laboratory-colonized specimen; LC+NE, laboratory-colonized and naturally exposed specimen; MPN, most probable number; TTC, 2,3,5triphenyltetrazolium chloride. * Corresponding author. Tel.: +44-1382-496341; fax: +44-1382-633952. E-mail address: [email protected] (S. Bartosch). 1 The work presented has been done at the University of Hamburg. The addresses given are the current addresses of the authors. 0167-7012/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 7 0 1 2 ( 0 2 ) 0 0 1 3 3 - 1
76
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
1. Introduction The contribution of microorganisms to the deterioration of natural stone, mortar and concrete has been widely investigated during the last 10 – 15 years (Griffin et al., 1991; Lyalikova and Petrushkova, 1991; Ortega-Calvo et al., 1991; Palmer et al., 1991; De la Torre et al., 1993; May et al., 1993; Hirsch et al., 1995a; Warscheid and Krumbein, 1996; Mansch and Bock, 1998; Warscheid and Braams, 2000). Chemoorganotrophic bacteria, fungi and chemolithoautotrophic nitrifying bacteria represent the epi- and endolithic microflora most relevant to the biodeterioration of natural building stone (Warscheid and Krumbein, 1996; Mansch and Bock, 1998). To evaluate the danger to historical buildings, monuments or statues from biodeterioration, it is essential to quantify these microorganisms. Enumeration is usually done by plate count and most probable number (MPN) techniques, which are based on the cultivation of microorganisms on selective media. These techniques need a relatively high amount of sample material, and are quite timeconsuming requiring 1 –6 weeks of incubation. Due to suboptimal culture conditions and methodological limitations, the cell numbers are usually underestimated. An overview of destructive and nondestructive methods used to study the stone-inhabiting microflora in rocks and building stones has been given by Hirsch et al. (1995b). In comparison to cultivation techniques, microscopic cell counts can be much faster for the evaluation of microbial colonization and usually higher cell numbers are found (Richaume et al., 1993). Confocal laser scanning microscopy (CLSM) offers the opportunity to visualize microorganisms in situ within the pore system of mineral materials (Quader and Bock, 1995; Bartosch et al., 2002). Fluorescence signals can be measured by the digital image analysis system which allows semi-automatic counting of microorganisms in stone material (Bartosch et al., 1996). However, many fluorescence dyes including DAPI which is widely applied to study bacteria in natural samples proved to be unsuitable, as these dyes bind to the stone material and cause strong background fluorescence which hinders visualization of the microorganisms (unpublished). Only staining by means of acridine orange (AO) has been shown to be suitable to visualize microorganisms on and in natural stone
(Quader and Bock, 1995). After dye application, the green fluorescing microorganisms can be easily distinguished from the red fluorescing mineral components (e.g. clay particles). However, with AOstaining, it is not possible to distinguish between active and inactive microorganisms, as they are all stained green, whereas after autoclaving the stone material, dead cells show red fluorescence. Thus, dead cells can hardly be distinguished from red fluorescing stone matrix. For assessment and monitoring of the risk of biodeterioration of a historical building, it may be necessary to know the in situ activity of microorganisms. Tetrazolium salts are commonly used to analyze the activity of natural populations. They act as artificial electron acceptors within functional electron transport systems (e.g. respiratory) or for certain active dehydrogenases. So far, little work has been done on activity analysis of stone-inhabiting microorganisms. The tetrazolium salts 2,3,5-triphenyltetrazolium chloride (TTC) (Warscheid et al., 1990) and 2-( p-iodophenyl)-3-( p-nitrophenyl)-5-phenyltetrazolium chloride (INT) (Taylor and May, 1995, 2000) have been used to investigate microbial activity in stone. These approaches were limited by the fact that the reduction products have to be extracted for quantitative study and this usually underestimates the cell numbers. In the study presented here, the tetrazolium salt 5cyano-2,3-ditolyltetrazolium chloride (CTC) has been used for the first time to visualize and quantify in situ the actively respiring microorganisms in natural stone. In previous studies, CTC had been applied successfully to quantify active bacteria in groundwater, fresh water and seawater (Rodriguez et al., 1992; Sherr et al., 1999), drinking water (Schaule et al., 1993; Kalmbach et al., 1997) activated sludge (Rodriguez et al., 1992; Griebe et al., 1995), estuarine sediments (Proctor and Souza, 2001) and soil (Winding et al., 1994; Yu et al., 1995). In contrast to INT and TTC, the formazan crystals formed by CTC reduction (CTF) show red fluorescence and CTC is therefore suitable for the visualization of microorganisms in stone using CLSM. To establish a method for selective counting of active microorganisms, representative pure cultures of stone-inhabiting microorganisms were checked for their ability to reduce CTC to CTF. The staining
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
conditions for microorganisms in natural stone were optimized and the influence of temperature and humidity on the microbial activity was studied. CTC staining was applied to determine the percentage of active bacteria in natural stone at an urban exposure site in Germany.
2. Materials and methods 2.1. Bacterial strains and culture conditions Pure cultures of Micrococcus varians ATTC 15306 and Paracoccus denitrificans ATTC 19367 were obtained from the German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany. All other strains used in this study had been isolated from building stone. The chemolithoautotrophic bacteria Nitrosovibrio spec. K7.1 and Nitrobacter vulgaris K55, as well as the facultatively methylotrophic bacterium Methylobacterium fuyisawaense L5, were isolated from Cologne Cathedral, Germany. The actinomycete Geodermatophilus spec. BC 532 was isolated from natural stone of the church Santa Maria Scala Quarry in Noto, Italy and kindly supplied by Dr. Clara Urzı´, University of Messina, Italy. The yeast Sporobolomyces roseus Kluyver and Niel, the filamentous fungi Phoma leveillei Boerema and Bollen, and Trichoderma citrinoviride Bissett were isolated from stone specimens at an exposure site in Duisburg, Germany. All strains are stored in the culture collection of the Institut fu¨r Allgemeine Botanik at the University of Hamburg, Germany. M. varians ATTC 15306, P. denitrificans ATTC 19367, and Geodermatophilus spec. BC 532 were cultivated on Standard I nutrient broth (Merck No. 107882). Nitrosovibrio spec. K7.1 was cultivated in a mineral medium with 10 mM NH4Cl (Koops et al., 1991). N. vulgaris K55 was grown lithoautotrophically in a mineral medium with 30 mM NaNO2 (Bock et al., 1990). M. fuyisawaense L5 was grown in a basal salt solution of pH 8.4 with 1.5% methanol according to Green et al. (1988). For cultivation of S. roseus Kluyver and Niel, P. leveillei Boerema and Bollen, and T. citrinoviride Bissett, a Sabouraud-Maltose agar (Merck No. 5439) was used. For CTC analysis, these fungi were transferred and grown in Malt extract broth (Merck No. 105397).
77
2.2. Stone samples and experimental conditions All investigations were done using Ihrlersteiner Green Sandstone. This type of sandstone is a dolomitic material, which supports the growth of chemoorganotrophic bacteria and lithoautotrophic nitrifying bacteria (Mansch and Bock, 1998). Stone samples 5 mm thick were taken from the stone specimens for investigation by using a diamond-coated cutting disc (Ø 20 0.6 mm thick) attached to a Proxxon hobbydriller. Stone specimens of size 1.5 1.5 1.5 cm were initially inoculated in the laboratory with pure cultures of ammonia and nitrite oxidizers and kept for several years in a plastic box on fine grain sand at maximum water holding capacity in an atmosphere of about 98% relative humidity and at 28 jC. Due to the nonsterile conditions, a complex microflora developed on these specimens. These specimens were used to determine the optimal CTC concentration and incubation time and are referred to as laboratory-colonized (LC) specimens. The influence of drying on the number of active bacteria was determined on the above-described LC specimens. During drying, the specimens were stored at 28 jC and 40% relative humidity and were analyzed after 0, 1, 2, and 5 days. After 5 days, the samples were remoistened with sterile water and analyzed after 6 h. To analyze the influence of temperature on the number of active bacteria, LC specimens of size 1.5 5 5 cm were used. These specimens had been additionally exposed for 2 years to outside conditions in the botanical garden of the Institut fu¨r Allgemeine Botanik at the University of Hamburg, Germany and are referred to as laboratory-colonized and naturally exposed (LC + NE) specimens. During this exposure algae had grown on the stone surface. Prior to CTC staining, the samples were incubated at different temperatures (5, 10, 28, and 35 jC) for 4 weeks at high stone moisture (maximum water holding capacity) and in an atmosphere of about 98% relative humidity. CTC staining on naturally colonized stone was carried out on samples taken from an urban exposure site in Duisburg, Germany on three different occasions in summer 1997. The site has been previously described by Mansch and Bock (1998). The specimen
78
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
sampled for this study was 20 30 45 cm and had been exposed for 9 years. Six samples were collected from vertical surfaces exposed to the east, west, north, and south side of the stone specimen, respectively, on 5th August 1997. At that time, it had not rained for 4 days and the daytime temperature ranged between 20 and 25 jC. During transport, the samples were kept in small containers to protect them from breaking and were sealed to avoid drying out of the natural moisture of the stone material. Transport took approximately 6 h and samples were stained immediately on arrival in the laboratory. 2.3. Staining with CTC Pure liquid culture samples (1 ml) were taken aseptically at different stages of growth (lag, exponential, and stationary phase). The growth phases were checked by microscopic cell counts using a Helber chamber for M. varians, P. denitrificans, M. fuyisawaense, S. roseus, N. vulgaris, and Nitrosovibrio spec. Cell protein was determined colorimetrically by using the method of Bradford (1976), as modified by Spector (1978) for Geodermatophilus spec., P. leveillei, and T. citrinoviride. About 100 Al of a 50 mM CTC solution (Polyscience Europe) was added to the samples and incubated for 2 h at room temperature in the dark. The cells were then washed with sterile 0.9% NaCl and suspended in 500 Al of the 0.9% NaCl solution, and 500 Al of a 4% formaldehyde solution was added to stop CTC reduction. After fixation for 10 min on ice, cells were washed twice with 0.9% NaCl. Counterstaining of total cells was done with 4,6,diamidino-2-phenylindol (DAPI) (10 Ag ml 1) for 5 min. Optimal CTC concentration and incubation time for staining of microorganisms in stone were tested. The following procedure achieved the best result. Freshly cut stone samples with a size of 5 5 5 mm were stained with 100 Al of a 15 mM CTC solution following a 24-h incubation period at room temperature in the dark. During incubation, the samples were kept in Eppendorf-tubes so that they were covered by the CTC solution. To stop CTC reduction, the stone samples were dried in air and stored at 20 jC. Control samples were treated with 2% formaldehyde or 0.1% sodium azide prior to CTC application. Staining of total cells was performed on parallel
samples with 200 Ag l 1 AO dissolved in distilled water containing 1% dimethylsulfoxid (Merck) for 1 min. 2.4. Microscopic determination of cell numbers Stained samples of pure cultures were filtered onto a polycarbonate filter with a pore size of 0.2 Am (Costar, Bodenheim) according to Hobbie et al. (1976). Cell numbers were determined with a Leica DM IRBE inverse microscope using a PL Fluotar objective (40 /1.0/oil) and an ocular L Plan (10 ). A Leica filter set N 2.1 (BP 515– 560 exc./RKP 580/ LP 590 em.) was used for CTF detection. DAPI was visualized with a Leica filter set A (BP 340– 380 exc./ RKP 400/LP 425 em.). Cell counts were carried out within a square of an area of 6.25 105 Am2 inside the ocular. At least 400 cells on 20 randomly chosen areas on a filter where counted and the mean value and standard deviation was calculated. Microscopic analysis of stone samples was done on aluminum slides prepared with a drill hole in the center to which a cover glass was fixed. The stained samples were placed in a drop of water on the cover glass and examined with an inverse microscope. Cell numbers were determined on transverse sections cut at 90j to the stone surface. On six parallel stone samples, five cell counts were done on different areas at various depths in the ranges 0 –0.5, 0.5– 2.0, and 2.0– 5.0 mm to achieve the mean cell number for a layer of 5-mm overall depth. The stone samples were investigated with a CLSM (Leica model TCS 4D) using a PL Fluotar objective (40x/1.0/oil). CTF was excited at 568 nm with an argon krypton laser and the emitted red fluorescence was detected with a long pass filter at 590 nm. AOstained microorganisms were excited at 488 nm and the emitted green fluorescence detected with a band pass filter at 520 – 560 nm. Simultaneous staining with CTC and AO is not possible as AO also shows red fluorescence when it is excited with the 568-nm wavelength and so CTC and AO-staining was done on different subsamples. Laser power, pinhole, and photo multiplier were adjusted to bright fluorescence of the microorganisms with the background showing no fluorescence. A series of 25 optical sections at a distance of 1 Am apart was recorded for each cell count. The total fluorescence intensity of the stone
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
sample was measured by means of the image analysis program (SCANware 5.1, Leica) within an area of 250 250 Am covering a depth of 25 Am. The mean fluorescence intensity of three single bacterial cells was measured and the cell number of the scanned volume unit was calculated as follows: cell number total fluorescence intensity within measured space ¼ mean bacterial fluorescence intensity of a single cell
Using these data, the cell numbers per gram of stone material were extrapolated.
3. Results Staining of pure cultures was achieved with 5 mM CTC, incubated for 2 h at room temperature. As shown in Table 1, chemoorganotrophic bacteria and fungi were able to reduce CTC to red fluorescing CTF, while lithoautotrophic nitrifying bacteria could not be stained. In the case of M. varians, P. denitrificans, M. fuyisawaense, and S. roseus, the percentage of CTF-positive organisms was correlated to the growth phase. During the lag phase, 15 –30% of total cells were stained. The percentages of CTF-positive cells increased during the exponential growth phase to 65 –79% of total cells. During the stationary phase, the number of CTF-positive cells decreased to 15– 28% of total cells. It was not possible to quantify
79
CTF-positive cells of Geodermatophilus spec., P. leveillei, and T. citrinoviride with the microscope because they are filamentous organisms. Therefore, the growth phase was determined by measuring calorimetrically the increase in cell protein. Microscopic observations indicated that CTF crystals accumulated in the cells at approximately the same rate at each stage of their growth phase. In contrast to pure cultures, optimal staining of bacteria on natural stone was achieved at a CTC concentration of 15 mM and after an incubation of 24 h. Single cells stained by CTC could be detected on LC specimens (Fig. 1a). Bacterial microcolonies were found on naturally colonized stone derived from the exposure site in Duisburg (Fig. 1b). Control samples treated with formaldehyde or sodium azide showed no fluorescence. When LC samples were dried, the number of CTFpositive cells decreased (Fig. 2), while the total cell numbers determined with AO remained constant (6.5 109 cells g 1 stone). Under wet conditions, 60% of total cells were stained with CTC. About 39% of total cells were still active after 1 day of drying, and after 2 days of drying, the percentage of CTF-positive cells decreased to 20%. This level did not change during three more days of drying. Remoistening of the samples after 5 days of drying resulted in an increase of active cells to 48% after 6 h. The influence of temperature on the number of active cells on LC + NE specimen is shown in Fig. 3.
Table 1 CTC reduction of representative stone-inhabiting (a) chemoorganotrophic bacteria, (b) fungi, and (c) nitrifying bacteria Organisms
(a)
(b)
(c)
Micrococcus varians Paracoccus denitrificans Methylobacterium fuyisawaense Geodermatophilus spec. Sporobolomyces roseus Phoma leveillei Trichoderma citrinoviride Nitrosovibrio spec. Nitrobacter vulgaris
CTC reduction
CTC reduction of sterilized cells
Percentage of CTF positive cells during the exponential phase
+ + + + + + +
66 – 78% 68 – 79% 68 – 79% n.a. 65 – 68% n.a. n.a. n.d. n.d.
Microscopic cell counts of CTF positive cells of Geodermadophilus spec., Phoma leveillei, and Tichoderma citronoviride could not be determined because they are filamentous organisms. n.a. = not analyzed. n.d. = not detected.
80
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
Fig. 2. Influence of drying (28 jC, 40% relative humidity) on the percentage of total (AO) and active (CTC) cells counted on laboratory-colonized specimens of Ihrlersteiner Green Sandstone incubated at high stone moisture (5R) (maximum water holding capacity). After remoistening content the activity of the cells recovered.
A maximum of 60% active cells was found at 28 jC, whereas at 5, 10, and 35 jC, the percentages of active cells were considerably lower than at 28 jC with values between 25% and 32% of CTC-stained cells. AO staining revealed that the total cell numbers stayed at a constant level of about 9.1 107 cells g 1 stone for all temperatures studied. On naturally colonized Ihrlersteiner Green Sandstone from the exposure site in Duisburg, 18 – 56% of the total cells were stained by CTC. The measured CTC reduction activity could be attributed mainly to bacteria because fungi were rarely detected during microscopic investigations. On the north-, east-, and
Fig. 1. Confocal laser scanning micrographs of CTC-stained bacteria on natural stone. (a) Single red fluorescing bacteria (B) distributed within the pore system of a laboratory-colonized specimen. The stone (S) material showed slight green autofluorescence. (b) Red fluorescing bacterial microcolony (M) visualized on natural stone colonized under natural conditions. Bars = 5 Am. CTC-stained cells and stone autofluorescence were visualized with a CLSM (microscope: Leica model TCS 4D); excitation was provided by an argon krypton laser.
Fig. 3. Influence of temperature on the percentage of active (CTC) and total (AO) cells on laboratory-colonized stone specimen, which had been exposed additionally to outside conditions. The specimens of Ihrlersteiner Green Sandstone had been stored for 4 weeks at different temperatures before staining.
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
Fig. 4. Active and total cell numbers at different sides of a specimen from Ihrlersteiner Green Sandstone exposed for 9 years at an urban exposure site in Duisburg, Germany. Sampling took place on 5th August 1997 after 4 days without rain and a daytime temperature ranging between 20 and 25 jC.
west-oriented sides, a nearly constant level of 52– 56% of total bacterial cells (average 6.5 107 cells g 1 stone) were CTF-positive (Fig. 4), whereas on the south side, only 18% of 6.7 106 cells g 1 stone were found to be active.
4. Discussion Pure culture experiments with selected strains belonging to the stone-inhabiting microflora showed that chemoorganotrophic bacteria and fungi were able to reduce the redox dye CTC to CTF. The optimal staining conditions were found to be 5 mM CTC incubated for 2 h (data not shown) which corresponds to previous findings (Griebe et al., 1995; Choi et al., 1999). For the chemoorganotrophic bacteria and S. roseus, the amount of cells reducing CTC was dependent on the growth phase, whereby cells from the exponential phase showed the highest activity similar to previous findings (Sherr et al., 1999). In contrast, the lithoautotrophic nitrifying bacteria could not be stained by CTC. CTC is reduced in the presence of functional electron transport (i.e. respiratory) systems or certain active dehydrogenases. Although ammonia and nitrite oxidizers possess high amount of electron transport proteins (Bock et al., 1991), we hypothesize that the redox potential of the electron carriers was too low to reduce a detectable amount of CTC under the given staining conditions. Increased incubation time
81
up to 24 h and CTC concentration up to 15 mM (data not shown) did not alter the results. However, under these conditions, toxic effects have to be taken into account as previously reported for liquid cultures and seawater samples (Griebe et al., 1995; Choi et al., 1999). We demonstrated that microorganisms in natural stone could be stained by CTC. Control samples treated with biocides (formaldehyde, azide) clearly showed that chemical CTC reduction can be excluded. Indeed, bacterial CTC reduction correlated with temperature which has already been shown to influence the activity of marina bacteria (Choi et al., 1999) and water stress. A similar correlation between bacterial activity, water stress, and temperature has also been observed for soil samples (Winding et al., 1994). These findings therefore provide clear evidence that the CTC reduction in stone was induced by microbial activity and the findings were also confirmed by the analysis of the specimens from the exposure site in Duisburg. Under natural conditions, the lowest numbers of active bacteria were found at the south side which is more exposed to sunshine. Thus, these microorganisms were more influenced by UV light and the microclimate in the stone was very likely characterized by increased water stress and intermittent high temperatures which can have a negative effect on microbial activity above 28 jC as demonstrated on the laboratory-colonized specimens (Fig. 3). On natural stone a much higher CTC concentration and longer incubation time was necessary for the staining of bacteria than in pure cultures. A CTC concentration of 15 mM obviously did not have toxic effects on the stone-inhabiting microorganisms, although toxic effects have been observed for liquid cultures at concentrations above 10 mM (Griebe et al., 1995). For seawater samples, a slight decrease of CTF positive cells was observed when 15 mM CTC was used compared to 5 mM CTC, but CTC reduction still took place (Choi et al., 1999). Previous investigations on soil samples showed that longer incubation times were required for successful staining compared to the conditions used for analysis of liquid cultures (Yu et al., 1995). Two factors may be responsible for this finding on stone material: (i) a certain percentage of CTC might be adsorbed at the pore surface of the mineral matrix; (ii) extracellular polymeric substances, which usually surround bacteria growing attached
82
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
to mineral surfaces, might have hindered the penetration of CTC into the cells. CLSM cell counts of microorganisms in stone were attributed mainly to the chemoorganotrophic bacterial microflora, whereas nitrifying bacteria, the main representatives of the lithoautotrophic stone-inhabiting microflora, and fungi hardly played a role during microscopic quantification. In fact, cell number determination of microorganisms in stone using CLSM is a fairly insensitive method with a detection limit of 105 cells g 1 stone (Bartosch et al., 1996) compared to plate or MPN counts, which have a detection limit between 101 and 102 CFU g 1 stone depending on the methodological approach. While chemoorganotrophic bacteria are frequently found in natural building stone using cultivation techniques at cell numbers above 105 CFU g 1, fungi and nitrifying bacteria are detectable at lower cell numbers by at least one or three orders of magnitude respectively (Mansch and Bock, 1998). In over 500 samples from historical buildings containing nitrifying bacteria, the average cell number was found to be 3.8 101 cells g 1 stone for ammonia-oxidizers and 5.3 101 cells g 1 for nitrite oxidizers (Mansch and Bock, 1998) which is far below the microscopic detection limit. In addition, fungi, which can be distinguished morphologically from bacteria, were very rarely observed during the microscopic investigations in this study. However, the ability of some stone-inhabiting fungi to form thick, melaninized cell walls (Wollenzien et al., 1995) may prevent the penetration of dyes and should also be taken into consideration. In the case of heavy growth of filamentous fungi, the semi-automated quantification of active bacteria by CLSM would certainly be obstructed because it is impossible to differentiate between CTF crystals accumulated in fungi or those accumulated in bacteria. Algae were found on the light-exposed surfaces of LC + NE specimens and the naturally colonized specimen from Duisburg. On vertical facades of historical buildings, algae are usually restricted to those areas which are in direct contact with running water or affected by rising damp. Algae and cyanobacteria can be easily detected without the application of fluorescence dyes as their chlorophyll shows red autofluorescence. However, this fluorescence spectrum is similar to that of CTF, a problem which concerns many commercially available fluorescence dyes (Lie-
sle et al., 1999). For investigations of complex biofilms containing algae and cyanobacteria, the autofluorescing cells could be discriminated from CTC-stained cells by separating the red CTC fluorescence from the far red fluorescence originating from the chlorophyll containing organisms (Neu, 2000). However, for biodeterioration studies, it is necessary to analyze the microorganisms growing within the depth of the building material. For this reason, bacteria were counted on transverse sections from 0- to 5-mm depth because previous studies using cultivation techniques revealed that bacteria and fungi are found mainly within this depth (Mansch and Bock, 1998). On the analyzed transversal sections, algae were not detected and bacterial cell counts could be done unhindered using the red spectrum of the fluorescence light. In conclusion, CTC staining in combination with CLSM quantification was proven to be a suitable method to determine the in situ activity status of chemoorganotrophic bacteria in natural stone. A selective method to determine the state of activity of nitrifying bacteria in natural building stone is well established and has been widely applied (Mansch and Bock, 1996, 1998). This is a different approach whereby the end products of ammonia and nitrite oxidation, i.e. nitrite and nitrate, are measured in suspensions of ground stone material using high performance liquid chromatography. However, due to the variety of substrates, a comparable technique has not been described for chemoorganotrophic bacteria. CTC staining in combination with CLSM imaging is a quick and reliable method for the evaluation of the current microbial in situ activity on historical buildings. Only small samples are required and this is of great importance in minimizing the damage to such valuable objects. Therefore, this method is very promising in being helpful in assessing the risk of biodeterioration of historical buildings or, for example, to monitor the efficacy of preventive measures such as treatment with biocides.
Acknowledgements This study was supported by the German Ministry of Research Education and Technology (BMBF) through funding of the projects Bau 5016E and Bau 7016E.
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
References Bartosch, S., Quader, H., Bock, E., 1996. Confocal laser scanning microscopy: a new method for detecting microorganisms in natural stone. DECHEMA Monogr. 133, 37 – 43. Bartosch, S., Mansch, R., Pinck, C., Bock, E., 2002. Visualisation and quantification of microorganisms on mortar on natural stone by CLSM. In: Gala´n, E., Zezza, F. (Eds.), Protection and Conservation of the Cultural Heritage of the Mediterranean Cities. A.A. Balkema Publishers, Lisse, pp. 275 – 279. Bock, E., Koops, H.-P., Mo¨ller, U.C., Rudert, M., 1990. A new facultatively nitrite oxidizing bacterium, Nitrobacter vulgaris sp. nov. Arch. Microbiol. 153, 105 – 110. Bock, E., Koops, H.-P., Harms, H., Ahlers, B., 1991. The biochemistry of nitrifying bacteria. In: Shively, J.M., Barton, L.L. (Eds.), Variations in Autotrophic Life. Academic Press, London, pp. 171 – 200. Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle protein-dye binding. Anal. Biochem. 72, 249 – 254. Choi, J.-W., Sherr, B.F., Sherr, E.B., 1999. Dead or alive? A large fraction of ETS-inactive marine bacterioplancton cells, as assessed by reduction of CTC, can become ETS-active with incubation and substrate addition. Aquat. Microb. Ecol. 18, 105 – 115. De la Torre, M.A., Go´mez-Alarco´n, G., Vizcaino, C., Garcia, T., 1993. Biochemical mechanisms of stone alteration carried out by filamentous fungi living in monuments. Biogeochemistry 19, 129 – 147. Green, P.N., Bousfield, I.J., Hood, D., 1988. Three new Methylobacterium species: M. rhodesianum sp. nov., M. zatmani sp. nov., and M. fuyisawanease. Int. J. Syst. Bacteriol. 38, 124 – 127. Griebe, T., Schaule, G., Secker, J., Flemming, H.-C., 1995. Bestimmung der stoffwechselaktiven Bakterien im belebten Schlamm. Wasser Abwasser 136, 608 – 615. Griffin, P.S., Indictor, N., Koestler, R.J., 1991. The biodeterioration of stone: a review of deterioration mechanisms, conservation case histories, and treatment. Int. Biodeterior. 28, 187 – 207. Hirsch, P., Eckhardt, F.E.W., Palmer, R.J., 1995a. Fungi active in weathering of rock and stone monuments. Can. J. Bot. 73, 1384 – 1390. Hirsch, P., Eckhardt, F.E.W., Palmer, R.J., 1995b. Methods for the study of rock-inhabiting microorganisms—a mini review. J. Microbiol. Methods 23, 143 – 167. Hobbie, J.E., Daley, R.J., Jasper, S., 1976. Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33, 1225 – 1228. Kalmbach, S., Manz, W., Szewzyk, U., 1997. Dynamics of biofilm formation in drinking water: phylogenetic affiliation and metabolic potential of single cells assessed by formazan and in situ hybridization. FEMS Microbiol. Ecol. 22, 265 – 279. Koops, H.-P., Bo¨ttcher, B., Mo¨ller, U.C., Pommerening-Ro¨ser, A., Stehr, G., 1991. Classification of eight new species of ammonia oxidizing bacteria: Nitrosomonas communis sp. nov., Nm ureae sp. nov., Nm aestuarii sp. nov., Nm marina sp. nov., Nm nitrosa sp. nov., Nm eutropha sp. nov., Nm oligotropha sp. nov. and Nm halophila sp. nov. J. Gen. Microbiol. 137, 1689 – 1699.
83
Liesle, J.T., Stewart, P.S., McFeters, G.A., 1999. Fluorescent probes applied to physiological characterization of bacterial biofilms. In: Doyle, R.J. (Ed.), Biofilms: Methods in Enzymology. Academic Press, New York, pp. 166 – 178. Lyalikova, N.N., Petrushkova, Y.P., 1991. Role of microorganisms in the weathering of minerals in building stone of historical buildings. Geomicrobiol. J. 9, 91 – 101. Mansch, R., Bock, E., 1996. Simulation of microbial attack on natural and artificial stone. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Microbially Influenced Corrosion of Materials. Springer-Verlag, Berlin, pp. 167 – 186. Mansch, R., Bock, E., 1998. Biodeterioration of natural stone with special reference to nitrifying bacteria. Biodegradation 9, 47 – 64. May, E., Lewis, F.J., Pereira, S., Taylor, S., Seaward, M.R.D., Allsopp, D., 1993. Microbial deterioration of building stone—a review. Biodeterior. Abstr. 7, 109 – 123. Neu, T.R., 2000. In situ cell and glycoconjugate distribution in river snow studied by confocal laser scanning microscopy. Aquat. Microb. Ecol. 21, 85 – 95. Ortega-Calvo, J.J., Hernandez-Marine, M., Saiz-Jimenez, C., 1991. Biodeterioration of building materials by cyanobacteria and algae. Int. Biodeterior. 28, 165 – 185. Palmer Jr., R.J., Siebert, J., Hirsch, P., 1991. Biomass and organic acids in sandstone of a weathering building: production by bacterial and fungal isolates. Microb. Ecol. 21, 253 – 266. Proctor, L.M., Souza, A.C., 2001. Method for enumeration of 5cyano-2,3-ditolyltetrazolium chloride (CTC)-active cells and cell-specific CTC activity of benthic bacteria in riverine, estuarine and coastal sediments. J. Microbiol. Methods 43, 213 – 222. Quader, H., Bock, E., 1995. Confocal-laser-scanning microscopy: a new method to study microorganisms colonizing natural stone. Int. J. Restor. Build. 1, 295 – 304. Richaume, A., Steinberg, C., Jocteur-Monrozier, L., Faurie, G., 1993. Differences between direct and indirect enumeration of soil bacteria: the influence of soil structure and cell location. Soil Biol. Biochem. 25, 641 – 643. Rodriguez, G.G., Phipps, D., Ishiguro, K., Ridgway, H.F., 1992. Use of a fluorescent redox dye for direct visualisation of actively respiring bacteria. Appl. Environ. Microbiol. 58, 1801 – 1808. Schaule, G., Flemming, H.-C., Ridgway, H.F., 1993. Use of 5-cyano-2,3-ditolyltetrazolium chloride for quantifying planctonic and sessile respiring bacteria in drinking water. Appl. Environ. Microbiol. 59, 3850 – 3857. Sherr, B.F., Del Giorgio, P., Sherr, E.B., 1999. Estimating abundance and single-cell characteristics of respiring bacteria via the redox dye CTC. Aquat. Microb. Ecol. 18, 117 – 131. Spector, T., 1978. Refinement of Comassie-blue method of protein quantification. Anal. Biochem. 86, 142 – 146. Taylor, S., May, E., 1995. A comparison of methods for the measurement of microbial activity on stone. Stud. Conserv. 40, 163 – 170. Taylor, S., May, E., 2000. Investigations of the localisation of bacterial activity on sandstone from ancient monuments. Int. Biodeterior. Biodegrad. 46, 327 – 333. Warscheid, T., Braams, J., 2000. Biodeterioration of stone: a review. Int. Biodeterior. Biodegrad. 46, 343 – 368. Warscheid, T., Krumbein, W.E., 1996. General aspects and selected cases. In: Heitz, E., Flemming, H.-C., Sand, W. (Eds.), Micro-
84
S. Bartosch et al. / Journal of Microbiological Methods 52 (2003) 75–84
bially Influenced Corrosion of Materials. Springer-Verlag, Berlin, pp. 273 – 295. Warscheid, T., Petersen, K., Krumbein, W.E., 1990. A rapid method to demonstrate and evaluate microbial activity on decaying sandstone. Stud. Conserv. 35, 137 – 147. Winding, A., Binnerup, S.J., Sørensen, J., 1994. Viability of indigenous soil bacteria assayed by respiratory activity and growth. Appl. Environ. Microbiol. 60, 2869 – 2875.
Wollenzien, U., De Hoog, G.S., Krumbein, W.E., Urzi, C., 1995. On the isolation of microcolonial fungi occurring on and in marble and other calcareous rocks. Sci. Total Environ. 167, 287 – 294. Yu, W., Dodds, W.K., Banks, M.K., Skalsky, J., Strauss, E.A., 1995. Optimal staining and sample storage time for direct microscopic enumeration of total and active bacteria in soil with two fluorescence dyes. Appl. Environ. Microbiol. 61, 3367 – 3372.