Vertical distribution in and isolation of bacteria from ... - Springer Link

2 downloads 0 Views 550KB Size Report
Susumu Takiil, Toshifumi Kondal, Akira Hiraishi', Genki I. Matsumoto2*, ... Tsudanuma 2-17-1, Narashino, Chiba Prefecture 275, Japan ..... Bergey's Manual.
Hydrobiologia 135, 15-21 (1986). O Dr W. Junk Publishers, Dordrecht. Printed in the Netherlands.

Vertical distribution in and isolation of bacteria from Lake Vanda: an Antarctic lake Susumu Takiil, Toshifumi Kondal, Akira Hiraishi', Genki I. Matsumoto2*, Tamio Kawano3 & Tetsuya Torii4 'Tokyo Metropolitan University, Department of Biology, Faculty of Science, Fukazawa 2-1-1, Setagayaku, Tokyo 158, Japan, 2Tokyo Metropolitan University, Department of Chemistry, Faculty of Science, Fukazawa 2-14, Setagaya-ku, Tokyo 158, Japan, 3oita University, Department of Chemistry, Faculty of Education, Dannohara 700, dita Prefecture 870-11, Japan and 4Chiba Institute of Technology, Tsudanuma 2-17-1, Narashino, Chiba Prefecture 275, Japan Keywords: vertical distribution, total bacterial number, bacterial biomass, Caulobacter, phototrophic bacteria, Antarctic lake

Abstract

Vertical distribution of bacteria in Lake Vanda, an Antarctic meromictic lake, was examined by the acridine orange epifluorescence direct count method. Total bacteria were 10"-lo5 cells. ml-I in the water at 55 m depth and above, and increased drastically to lo7 ~ e l l s - m l -in~ the bottom water. Filamentous or long rodshaped bacteria occurred at a high frequency in the upper layers, but in the bottom layers most bacteria were coccoidal or short rods. Mean bacterial cell volume in water of between 10 m and 60 m deep was fairly large compared with common bacterial populations in seawater and lake water. Aerobic heterotrophic bacteria were recovered from the water of a depth of 30 m and above, and were assumed to belong to Caulobacter. Viable heterotrophic bacteria were not recovered from the high salinity deep water by media prepared with the same deep water. Phototrophic purple non-sulphur bacteria were isolated by enrichment cultures from water at 55 m depth. Introduction

Among a number of saline lakes and ponds in Antarctica, Lake Vanda is one of the best known. It is meromictic and is located in the depression of Wright Valley in the Dry Valleys region of south Victoria Land (77" 32'S, 161" 34'E). The deep water of the lake has a high salinity, different in composition from seawater. The major constituent is calcium chloride (Angino & Armitage, 1963; Torii et al., 1975). The temperature in the deep water layers is as high as 25 "C throughout the year, despite a mean atmospheric temperature of about -20 "C. This lake ecosystem has attracted much attention. Its biology is well studied (Wright & Burton, 1981), * Present address: The University of Tokyo, Department 'of Chemistry, College of Arts and Sciences, Komaba, Meguro-ku, Tokyo 153, Japan.

but information on its bacterial population, especially bacterial biomass, is as limited as in other Antarctic lakes and ponds (Wright & Burton, 1981). Meyer et al. (1962), Benoit et al. (1971) and Kriss et al. (1976) enumerated bacteria and tried to isolate some bacteria, and Vincent et al. (1981) estimated microbial activity in Lake Vanda. The acridine orange epifluorescence direct count (AODC) method is considered a reliable method for estimating aquatic bacterial numbers. This method has not been used much in the study of Antarctic lakes. We report here preliminary data on the vertical distribution of bacteria, estimated by the AODC method, in Lake Vanda. In addition we were able to isolate some bacteria. Bacterial counts were also carried out for deep waters in the east and west lobes of Lake Bonney (77" 4 3 5 , 162" 25'E; 28 km southeast of Lake Vanda), situated in Taylor Valley.

Materials and methods After drilling into lake ice using a Sipre ice auger, water samples were collected from several depths by a Kitahara-type water sampler on January 2, 1984 at the deepest point of Lake Vanda. Bottom waters were taken in the same way on December 10, 1983 in the east lobe and December 12, 1983 in the west lobe of Lake Bonney. Autoclaved polyethylene bottles (50 ml) previously rinsed with distilled water filtered through 0.45 pm Millipore filters were used in sampling. The samples were fixed in situ by addition of glutaraldehyde to give a final concentration of 1% (V/V). Water samples for recovery of bacteria were stored in autoclaved polyethylene bottles (100 ml). All samples were kept at about 5 "C during transport to our laboratory in Tokyo until examination. The total number of bacteria in the fixed samples were determined by the AODC method with an epifluorescence microscope according to Hobbie et al. (1977). Mean bacterial cell volume was estimated from measurement of about 50 cells per sample using epifluorescence microphotographs of cells on 0.2 pm Nuclepore filters used for direct counts. The factor for conversion of bacterial volumes to biomass (207.5 fg C , ~ ~ r nreported -~) by Watson et al. (1977) was employed. Detection of viable heterotrophic bacteria was carried out using two agar media. Medium A consisted of proteose peptone (Difco) 2.0 g . I-', polypeptone (Daigo) 2.0 g -1-', yeast extract (Difco) 1.0 g .I-', glucose 0.5 g . I-' and agar 15.0 g .I-'. Medium B was made by dilution of medium A to one-tenth fold except for agar. The water samples were diluted 10 fold in series with sterilized lake water collected from the same depth from which the water sample was obtained. An appropriate volume (0.1 ml) of suitable dilutions was spread on the agar plates. The inoculated plates were incubated in air or in GasPak anaerobic jars (BBL) at 20 "C for one month before counting. Aerobically grown colonies were picked and purified by repeated streaking on medium A plates. Characterization of isolates was carried out according to Hiraishi & Kitamura (1984). Detection of sulphate-reducing bacteria was attempted using a method of the most probable number (MPN) and anaerobic plates according to Tezuka (1979) and Wakao & Furusaka (1972).

Phototrophic purple sulphur bacteria and purple nonsulphur bacteria were enumerated according to Siefert et al. (1978) and Hiraishi & Kitamura (1984). All media described above were prepared with 75% (final concentration) lake water collected from the same depth as that of the water sample to be inoculated. In addition, an enrichment culture for photosynthetic bacteria was carried out as follows: Water samples of Lake Vanda from depths of 50 m and 60 m were diluted five-fold with sterilized distilled water, and yeast extract (Difco) added to give a final concentration of 0.1%. The solution was incubated for about two months at 30 "C under incandescent illumination (c. 2000 lux). Isolation, characterization and identification were carried out according to Triiper & Pfennig (1981) and Hiraishi & Kitamura (1984).

Results and discussion General qualities of lake water Water temperature in Lake Vanda increased stepwise with depth, from 0 "C at 0 m to 23.5 "C in the bottom water as shown in Fig. 1. Chlorinity was relatively low at a depth of 50 m and above, but abruptly increased below this depth and attained 76.5 g C1- kg-'. Dissolved oxygen was supersaturated at a depth of 55 m and above but then decreased rapidly, and anoxic conditions appeared in water below a depth of 65 m. There was a strong odor of hydrogen sulphide in the anoxic water samples. The vertical profiles were similar to those of previous surveys (Angino & Armitage, 1963; Torii et al., 1975; Matsumoto et al., 1982). In Lake Bonney, water temperature decreased below zero. It was 0 "C at a depth of 14 m and was -4.6 "C in the deep waters sampled for bacterial examination in both the lobes. The deep waters contained 2.97 ml O2 1 and 155.8 g C1- kg-' in the east lobe (34.6 m depth) and 0.01 ml O2 and 80.99 g C1 kg-' in the west lobe (32.7 m depth). No hydrogen sulphide was found in the bottom waters of Lake Bonney.

- -'

Direct counts and biomass of bacteria Figure 2 shows the vertical distributio~nof total

AODC (cells .ml-' ) Total cell volume ( J J

Dissolved oxygen iml .I-') I

0

-

1

1

I

I

I

I

60 20 40 Chlorinity ( g .kg-')

I I

1

80

0

I

1

I

1

~ ml-' ~ . I

) 1

0.2 (IL 0.6 0.8 1.0 1.2 Average cell volume lPm3. cell "1

Fig. I. Vertical profile of physicochemical properties of Lake Vanda.

Fig. 2. Vertical distribution of total bacteria by the AODC method and their average and total cell volumes in Lake Vanda.

bacteria measured by the AODC method in Lake Vanda on January 2, 1984. The direct counts were 1.0 x 105 cells. ml-' at a depth of 10 m and in the order of lo4 cells. ml-I in depths from 30 to 55 m. These values are somewhat lower than those for open seawater (Kogure et al., 1979) and oligotrophic lake waters (Aizaki et al., 1981). The direct counts increased drastically in the water below a depth of 55 m and reached 8.6 x lo6 cells ml-' at a depth of 68.5 m. Total bacterial numbers in Lake Vanda have not been reported, except that Kriss et al. (1976) enumerated bacteria by the capillary microscopy method. Our value for the water of 30 m depth was about one fourth of their estimate for the same depth reported to be the highest in the lake in their study. Our value for water of 60 m depth is about 5.4-fold greater than their value. In

the deep waters of Lake Bonney, the total counts and were 9.5 x 104 cells . ml-l, 2.5 x lo5 cells. ml-' at a depth of 34.6 m in the east lobe and 32.7 m in the west lobe, respectively. These values are much higher than those reported by Goldman et al. (1967). The differences between our study and others may be due to the enumeration methods used. We obtained reliable estimates of bacterial population numbers with the AODC method, even in the bottom waters of the lakes which contained a high concentration of particulate matter (Fig. 3b). However, the AODC method cannot distinguish between bacteria and cyanobacteria in a similar size range. The presence of small cyanobacteria such as Synechocystis and Phormidium in Lake Vanda has been reported by Goldman et al. (1967), but their

Fig. 3. Epifluorescence photomicrographs of acridine orange stained cells in the water samples from depths of 10 m (a) and 65 m (b). Amorphous particulate matters are seen at right and left corners in (b). Bar marker represents 5 pm.

numbers were below lo3 cells. ml-', less than one hundredth of our values. The total bacterial counts presented here, except those for the deep waters of Lake Vanda, are somewhat lower than those in other Antarctic lakes, for example lo5- lo6 cells ml-I in Lake Hoare (Mike11 et al., 1984) and lo6 - 10' cells . ml in Ace Lake (Hand & Burton, 1982), but fall within the range of 6.5 x lo4 to 6.5 x lo5 cell - ml-I in surface seawaters in McMurdo Sound in 1977 (Hodson et al., 1981). In Lake Vanda, the morphology of bacterial cells in the upper layer differed greatly from those in the bottom layer. The upper layer often contained filaments and long, rod-shaped bacteria, as shown in Fig. 3a. On the other hand, most bacteria from the bottom water were coccoidal, or short rods less than 1.0 pm long (Fig. 3b). The morphological properties of bacteria in the bottom waters of the east and west lobes of Lake Bonney were similar to those of the bottom water of Lake Vanda. Mean bacterial cell volumes are presented in Table 1. Bacteria from water of between 10 m and 60 m deep in Lake Vanda had relatively large mean cell sizes compared with bacteria from seawater off North Carolina and California, USA (Ferguson & Rublee, 1976; Fuhrman, 1981) and lake water of Lake Mendota, USA (Pedros-Alio & Brock, 1982). The cell volume ranged from 0.08 to

-'

0.3 pm3 . cell-' in the above cases. The total cell volume of bacteria in Lake Vanda showed a vertical profile similar to that of direct counts (Fig. 2). Bacterial biomass estimated from cell volumes (Table 1) ranged from 0.034 mg C .l-' (40 m depth) to 0.271 mg C - 1 - I (68.5 m depth), corresponding to from 0.22% (65 m depth) to 3.8% (10 m depth) of total organic carbon reported by Matsumoto et al. (1979) for samples collected on December 13, 1976. The low proportion of biomass to total organic carbon in the bottom layers may be caused by high concentrations of refractory organic matter (Matsumoto et al., 1984). The values of biomass obtained were extremely high compared to those for sea ice and for seawater of McMurdo Sound (Sullivan & Palmisano, 1984).

Isolation and characterization of bacteria Viable aerobic heterotrophic bacteria were about lo3 colony forming units (CFU).ml-' in the stored water samples from depths of 10 to 30 m. The number decreased to 10 CFU . ml-' in the water of 40 m depth and less than two CFU . ml in waters from a depth of 50 m and below. No significant difference in CFU's was found between media A and B. The anaerobically incubated plates gave rise to two or three CFU .ml-' only for water samples from depths of 60 m and 65 m. From the

-'

Table 1 . Bacterial cell volume and biomass in Lakes Vanda and Bonney. Average bacterial cell volume (pm'. cell- I)

Total bacterial cell volume (pm3.ml- I )

Lake Bonney, east lobe 34.6

0.257 k 0.254

(60)

2.5 x lo4

Lake Bonney, west lobe 32.7

0.180f 0.172

(54)

4.5 x 104

Water depth (m)

Bacterial biomass (mg C.1-I)

Total organic carbon** (mg C.1-I)

0.0058

15****

Lake Vanda 10 30 40 50 55 60 65 68.5

* Standard deviation. **After Matsurnoto et al. (1979). measured.

*** 33.4 m. **** 30.4 m. The figures in parentheses are the number of cells

deep waters of Lake Bonney, no viable bacteria were recovered. These numbers might perhaps not accurately reflect in situ viable counts of the water, because it took two weeks for the samples of Lake Vanda to be transported before inoculation in the laboratory in Tokyo. In fact, direct counts of the water samples used for inoculation were 1.4-5.0 times higher than those of water samples fixed in situ. The failure to detect bacterial colonies in deeper water may be a result of the death of bacteria during storage or from use of unsuitable media containing high salt concentrations. Studies of Lake Vanda and other Antarctic saline lakes and ponds reported the absence of recovery of bacteria on agar media prepared with in situ lake water of high salinity or seawater (Meyer et al., 1962; Benoit et a/., 1971; Cameron et al., 1972; Kriss et al., 1976; Waguri, 1976; Hand, 1980). Bacteria from Deep Lake could also not be cultured on media in which salinity was more than half that of the lake itself (Hand, 1980). Thus, Hand suggested that bacteria present might have been washed in from the surrounding catchment area and were not metabolically active in situ. On the other hand, Vincent et al. (1981) assayed various microbial activities of water samples from depths from 40 to 68 m in Lake Vanda and reported active DNA synthesis (3Hthymidine incorporation into DNA) at a depth of about 58 m just above the anoxic layer. All of the aerobically-grown colonies on agar plates were yellow and morphologically similar. Twenty bacterial strains were isolated by repeated streaking from plates obtained from depths of 10 m and 30 m. All isolates showed identical morphological and physiological properties, as follows: rodshaped cells, 2.5 - 10 pm by 0.6 - 1.0 pm, stalk formation, rosette formation (rare), no endospore formation, no branching filaments, Gram-negative, catalase weakly positive, cytochrome oxidase positive, no anaerobic growth, yellow pigmented, no bacteriochlorophyll. The isolates had motility at the initial stage of growth. In later stages they became rods of 5 to 10 pm length, resembling the long cells in situ. The isolates resembled Caulobacter according to Bergey's Manual, 8th edn. (Buchanan & Gibbons, 1974), and stalked bacterium, identified as Caulobacter has also been isolated from Lake Fryxell, a saline lake in the Dry Valleys (Waguri, 1976). Sulphate reducing bacteria, phototrophic purple

bacteria, or purple nonsulphur bacteria were not recovered in MPN tubes, agar shake tubes, or anaerobic plates, even with inocula of one ml in volume. Kriss et al. (1976) also could not isolate sulphate reducing bacteria using Postgate's medium B with 8% NaCl. Nevertheless, the presence of sulphate reducing bacteria is suggested by the presence of large amounts of hydrogen sulphide in the bottom waters of Lake Vanda (Torii et al., 1975). Further efforts are needed to isolate these bacteria. Enrichment cultures of water samples from depths of 55 m and 60 m gave a red color after two months of incubation in light. From the 55 m culture, photosynthetic bacteria were recovered and identified as Rhodomicrobiurn vannielii, Rhodopseudomonas palustris and Rhodospirillum spp. These bacteria might have grown heterotrophically in the oxic layer. All the isolates showed a sulphide tolerance of 1 to 2 mM. These photosynthetic bacterial species were also isolated by Herbert (1976) from sediment samples of three fresh water lakes and offshore coastal marine sediments in the Antarctic. Kriss et al. (1976) detected photosynthetic bacteria resembling the Chromatiurn group in sediment samples of Lake Vanda. In conclusion, total bacterial numbers estimated by the AODC method were considerably lower in the water above a depth of 60 m but increased drastically in bottom water, as much as 10' cells .ml-'. On the other hand, the average bacterial cell volume was considerably larger in the water above a depth of 60 m compared to that of the bottom water. It is unusual that an oligotrophic upper water contains bacteria with relatively large cells. In contrast, the bottom water with high concentration of organic materials contained smallsized bacteria. This suggests a change in species composition between upper and bottom water bacterial populations. From the water samples at depths of 10 m and 30 m, isolates belonging to Caulobacter were isolated. No recovery was obtained of viable heterotrophic bacteria from deep water with a high salinity; this corroborates earlier investigations.

Acknowledgment The authors express their gratitude to the Antarctic Division, DSIR, New Zealand for support in Antarctic research.

References Aizaki, M., A. Otsuki, T. Fukushima, M. Hosomi & K. Muraoka, 1981. Application of Carlson's trophic state index to Japanese lakes and relationships between the index and other parameters. Verh. int. Ver. Limnol. 21: 675-681. Angino, E. E. & B. Armitage, 1963. A geochemical study of Lakes Bonney and Vanda, Victoria Land, Antarctica. J. Geol. 71: 89-95. Benoit, R., R. Hatcher & W. Green, 1971. Bacteriological profiles and some chemical characteristics of two permanently frozen Antarctic lakes. In J. Cairns, Jr. (ed.), The Structure and Function of Fresh-Water Microbial Communities. Virginia Polytechnic and State University, Blacksburg: 281-293. Buchanan, R. E. & N. E. Gibbons (eds), 1974. Bergey's Manual of Determinative Bacteriology, 8th Edn. Williams & Wilkins Co., Baltimore: 153- 155. Cameron, R. E., E A. Morelli & L. P. Randall, 1972. Aerial, aquatic and soil microbiology of Don Juan Pond, Antarctica. Antarct. J. US 7: 452-458. Ferguson, R. L. & P. Rublee, 1976. Contribution of bacteria to standing crop of coastal plankton. Limnol. Oceanogr. 21: 141- 144. Fuhrman, J. A., 1981. Influence of method on the apparent size distribution of bacterioplankton cells: epifluorescence microscopy compared to scanning electron microscopy. Mar. Ecol. Prog. Ser. 5: 103-106. Goldman, C. R., D. T. Mason & J. E. Hobbie, 1967. Two Antarctic desert lakes. Limnol. Oceanogr. 12: 295-310. Hand, R. M., 1980. Bacterial populations of two saline Antarctic lakes. In P. A. Trudinger & M. R. Walter (eds.), Biogeochemistry of Ancient and Modern Environments. Proc. 4th int. Symp. Envir. Biogeochem. (ISEB): Springer-Verlag, Berlin: 123- 129. Hand, R. M. & H. R. Burton, 1982. Microbial ecology of an Antarctic saline meromictic lake. Hydrobiologia 82: 363 - 374. Herbert, R. A,, 1974. Isolation and identification of photosynthetic bacteria (Rhodospirillaceae) from Antarctic marine and freshwater sediments. J. appl. Bacterial. 41: 75-80. Hiraishi, A. & H. Kitamura, 1984. Distribution of phototrophic purple nonsulfur bacteria in activated sludge systems and other aquatic environments. Bull. jap. Soc. scient. Fish. 50: 1929- 1937. Hobbie, J. E., R. J. Daley & S. Jasper, 1977. Use of Nuclepore filters for counting bacteria by fluorescence microscopy. Appl. envir. Microbiol. 33: 1225-1228. Hodson, R. E., F. Azam, A. F. Carlucci, J. A. Fuhrman, D. M. Karl & 0. Holm-Hansen, 1981. Microbial uptake of dissolved organic matter in McMurdo Sound, Antarctica. Mar. Biol. 61: 89 - 94. Kogure, K., U. Simidu & N. Taga, 1979. A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25: 415-420. Kriss, A. E., I. N. Mitskevich, E. P. Rozanova & L. Osnitskaya, 1976. Microbiological investigations of Lake Vanda (Antarcti-

ca). Microbiology 45: 917-922. [Translated from Mikrobiologiya 45: 1075- 10811. Matsumoto, G., T. Torii & T. Hanya, 1979. Distribution of organic constituents in lake waters and sediments of the McMurdo Sound region in the Antarctic. In T. Nagata (ed.), Proceedings of the Seminar I11 on Dry Valley Drilling Project, 1978. Mem. natn. Inst. Polar Res., Spec. Issue 13. National Institute of Polar Research, Tokyo: 103-120. Matsumoto, G., T. Torii & T. Hanya, 1982. Nutrient matters in saline lakes of McMurdo Oasis in the 1976-77 summer season. Antarct. Rec. 74: 109-118. Matsumoto, G., T. Torii & T. Hanya, 1984. Vertical distribution of organic constituents in an Antarctic lake: Lake Vanda. Hydrobiologia 111: 119-126. Meyer, G. H., M. B. Morrow, 0. Wyss, T. E. Berg & J. Littlepage, 1962. Antarctica: the microbiology of an unfrozen saline pond. Science 138: 1103- 1104. Mikell, A. T., B. C. Parker & G. M. Simmons, Jr., 1984. Response of an Antarctic lake heterotrophic community to high dissolved oxygen. Appl. envir. Microbiol. 47: 1062-1066. Pedros-Alio, C & T. D. Brock, 1982. Assessing biomass and production of bacteria in eutrophic Lake Mendota, Wisconsin. Appl. envir. Microbiol. 44: 203-218. Siefert, E., R. L. Irgens & N. Pfennig, 1978. Phototrophic purple and green bacteria in a sewage treatment plant. Appl. envir. Microbiol. 35: 38-44. Sullivan, C. W. & A. C. Palmisano, 1984. Sea ice microbial communities: distribution, abundance, and diversity of ice bacteria in McMurdo Sound, Antarctica, in 1980. Appl. envir. Microbiol. 47: 788 - 795. Tezuka, Y., 1979. Distribution of sulfate-reducing bacteria and sulfide in aquatic sediments. Jap. J. Ecol. 29: 95-102. Torii, T., N. Yamagata, S. Nakaya, S. Murata, T. Hashimoto, 0. Matsubaya & H. Sakai, 1975. Geochemical aspects of the McMurdo saline lakes with special emphasis on the distribution of nutrient matters. In T. Torii (ed.), Geochemical and Geophysical Studies of Dry Valleys, Victoria Land in Antarctica. Mem. natn. Inst. Polar Res., Spec. Issue 4. National Institute of Polar Research, Tokyo: 5-29. Triiper, H. G. & N. Pfennig, 1981. Characterization and identification of the anoxygenic phototrophic bacteria. In M. P. Starr et a/. (eds.), The Prokaryotes, 1. Springer-Verlag,Berlin: 299-312. Vincent, W. F.,M. T. Downes & C. L. Vincent, 1981. Nitrous oxide cycling in Lake Vanda, Antarctica. Nature 292: 618-620. Waguri, O., 1976. Isolation of microorganisms from salt lakes in the Dry Valley, Antarctica, and their living environment. Antarct. Rec. 57: 80-96. Wakao, N. & C. Furusaka, 1972. A new agar plate method for the quantitative study of sulfate-reducing bacteria in soil. Soil. Sci. PI. Nutr. 18: 39-44. Watson, S. W., T. J. Novitsky, H. L. Quinby & F. W. Valois, 1977. Determination of bacterial number and biomass in marine environments. Appl. envir. Microbiol. 33: 940-946. Wright, S. W. & H. R. Burton, 1981. The biology of Antarctic saline lakes. Hydrobiologia 82: 319-338.