Growth Strategy of Heterotrophic Bacterial Population

Folia Microbiol. 50 (5), 427–435 (2005)

http://www.biomed.cas.cz/mbu/folia/

Growth Strategy of Heterotrophic Bacterial Population Along Successional Sequence on Spoil of Brown Coal Colliery Substrate V. KRIŠTŮFEKa, D. ELHOTTOVÁa, A. CHROŇÁKOVÁa,b, I. DOSTÁLKOVÁb, T. PICEKb, J. KALČÍKa aInstitute of Soil Biology, Academy of Sciences of the Czech Republic, 370 05 České Budějovice, Czechia

e-mail [email protected] bFaculty of Biological Science, South Bohemia University, České Budějovice, Czechia

Received 23 June 2005 Revised version 14 October 2005

ABSTRACT. The bacterial population of brown coal colliery spoil (Sokolov coal mining district, Czechia) was characterized by measuring viable bacterial biomass, the culturable to total cell ratio (C : T), colony-forming curve (CFC) analysis and species and/or biotype diversity. Bacterial representatives that differed in colony-forming growth (fast and/or slow growers) were used for growth-strategy investigation of heterotrophic bacteria. Spoil substrates from the surface (0–50 mm) and the mineral (100–150 mm) layers were sampled on 4 sites undergoing spontaneous succession corresponding to 1, 11, 21 and 43 years after deposition (initial, early, mid and late stages). The bacterial biomass of the surface layer increased during the initial and early stages with a maximum at mid stage and stabilized in the late stage while mineral layer biomass increased throughout the succession. The maxima of C : T ratios were at the early stage, minima at the late stage. Depending on the succession stage the C : T ratio was 1.5–2 times higher in the mineral than the surface layer of soil. An increase in the fraction of nonculturable bacteria was associated with the late succession stage. CFC analysis of the surface layer during a 3-d incubation revealed that the early-succession substrate contained more (75 %) rapidly colonizing bacteria (opportunists, r-strategists) than successively older substrates. The culturable bacterial community of the mineral layer maintained a high genera and species richness of fast growers along the succession line in contrast to the surface layer community, where there was a maximum in the abundance of fast growers in the early stage. There was a balanced distribution of Grampositive and Gram-negative representatives of fast growers in both layers. A markedly lower abundance of slow growers was observed in the mineral in contrast to the surface layer. Gram-positive species dominated the slow growers at the surface as well as in the mineral layers. The growth strategy of the heterotrophic bacterial population along four successional stages on spoil of brown coal colliery substrate in the surface layer displayed a trend indicative of a r–K continuum in contrast to the mineral layer, where an r-strategy persisted.

Soil formation is crucial to restoring the ecosystem function in post-mining landscapes. Brown colliery spoil sites offer the opportunity to investigate several stages of microbial succession during this soil formation. These gradients can change the growth strategy of organisms contributing to the succession, the r–K continuum (Gadgil and Solbrig 1972). The terms r and K have been applied to microbial successional theory by Andrews and Harris (1986) and later Atlas and Bartha (1993). They propose that as communities develop, opportunistic species (r-strategists) which invest a lot of energy in reproduction and have large niche widths are replaced by equilibrium species (K-strategists) which invest more energy in maintenance and have narrower niche widths. Since the ability to grow quickly on a nonselective agar medium could be considered as a characteristic of opportunistic bacteria, culturability and the percentage of culturable cells may be an indicator of the successional state in microbial populations. This concept was successfully used to compare bacterial populations (i) on wheat roots and soil in both field and glasshouse microcosms (De Leij et al. 1993), (ii) on maize roots (Chiarini et al. 1998), (iii) in laboratory microcosms of the rhizosphere of hydroponically grown wheat plants and aerobic, continuously stirred tank reactors (Garland et al. 2001) and (iv) in material exposed by a receding glacier (Sigler et al. 2002; Sigler and Zeyer 2004) and an extreme aridity (Jianping Su et al. 2004). This study tested the importance of culturability in heterotrophic bacterial population succession (r vs. K strategies) on a spoil of brown coal colliery substrate by measuring culturable to total cell ratios and colony-forming curves; these bacterial species and biotypes were also characterized over this primary succession.

428

V. KRIŠTŮFEK et al.

Vol. 50

2005

HETEROTROPHIC BACTERIAL POPULATION ON BROWN COAL SPOILS 429

430


Vol. 50

2005


MATERIALS AND METHODS Site and soil description. The study was carried out in the post-mining landscape of North-West Czechia (Sokolov coal mining district) on Great Podkrušnohorská Dump established on spoil from open-cast coal mining (Frouz and Nováková 2004). The original spoil substrate is mainly composed of tertiary cypris clay (pH 8) and represents the wastes of brown coal mining (for the properties of Sokolov-basin-sediments see Kříbek et al. 1998). Four sites (≈1 hm2), each undergoing spontaneous succession, represent four succession stages of a chronosequence (Table I). The initial stage (1 year after deposition) was freshly heaped spoil substrate of compact nonweathered clay stone. On the 11-year-old site (early stage), there was a sparse distribution of herbs and grasses (Tusilago farfara and Calamagrostis epigeios) and no effect of soil invertebrates could be seen on the microstructure of deposited clay. The transition from the 11 to 21-year-old site (mid stage) was marked by a rapid development of herbs, grasses and shrubs (Salix caprea). An accumulation of leaf litter and millipede excrements produced a dense fermentation layer (41 mm thick, see Table I). The 43-year-old site (late stage) was covered by two tree species: Populus tremuloides and Betula sp. Soils contained a well developed humus layer (63 mm) while the fermentation layer was thinner than that of the mid-stage. The main difference between the layers is that there is less total C, N and root mass in the mineral layer. These differences increased with succession age. The mineral layer did not contain fermentation and humus layers. Sampling. Samples (in three independent replicates, each from 13 sub-sampling places) were taken using a cylindrical soil corer (cross section area 3630 mm2) from the surface (0–50 mm) and the mineral layer (100–150 mm) in depressions of heap ribs at the end of dormancy in March 2004. Samples were homogenized by sieving through a 5-mm screen sieve and stored before analysis for 1–2 d at 4 °C. Estimation of bacterial biomass. Phospholipid fatty acids typical of bacteria (PLFAbact: i15:0, a15:0, 15:0, i16:0, cis16:1ω7, i17:0, cy17:0, cis18:1ω7 and cy19:0) were measured to estimate the total active bacterial community biomass (Frostegård and Bååth 1996). The branched fatty acids (i12:0, i13:0, i14:0, i15:0, a15:0, i16:0, i17:0, a17:0) were chosen to represent G+ bacteria (O’Leary and Wilkinson 1988), the cyclopropyl- and hydroxy-fatty acids (cy17:0, cy19:0, 2-OH10:0, 3-OH16:0, 3-OH16:1, 2-OH18:0) to represent G– bacteria (Wilkinson 1988; Zelles et al. 1992). PLFA identification methods were used according to Oravecz et al. (2004); the only modifications were that the final gas chromatographic detection step was with GC Agilent 6850, FID, Ultra 2 Column (cross-linked 5 % phenylmethyl siloxane; 25 m, 0.22 mm, 0.33 mm), H2 was the carrier gas and the temperature regime was 170 °C, 5 K/min, 260 °C 18 min, 40 K/min, 310 °C 1.5 min. Culturing. Triplicate soil samples (5 g wet mass) were suspended in 45 mL of 0.2 % solution of calgon (hexasodium hexametaphosphate), homogenized in an ultrasonic bath (50 kHz, 4 min). Samples were serially diluted (104–106) and plated (0.2 mL) in quadruplicate onto R2A agar Difco (pH 7.2) to estimate the relative proportion of the culturable cell population (C); plates were incubated in the dark at 20 °C. A colony-forming curve (CFC) was generated for each soil sample by counting newly visible colonies every day for a 7-d incubation period and plotting the cumulative number of colonies at each time point. Only plates containing 30 to 300 colonies were enumerated. “Fast growers” (r-strategists) were defined as bacteria that produce visible colonies within 3 d, “slow growers” (K-strategists) within 4–7 d. Total bacterial counts. The total number of bacteria (T) in each sample was estimated with DAPI (4´,6-diamidino-2-phenylindole) staining and microscopic counting (Bloem 1995). The ratio of culturable cells to total cells (C : T) was calculated to determine the index of succession state of microbial communities. Isolation and identification of fast- and slow-growing culturable bacteria. Bacterial colonies were isolated every day within the incubation period for consequent species characterization of the fast- and slowgrowers. Bacterial isolates were grouped according to incubation time and colony phenotype and purified by streak plating onto tryptic soy agar (Becton Dickinson Co., USA). Isolates were identified with whole cellular fatty-acid methyl esters (FAME) profiles using a MIS Sherlock automatic identification system (MIDI Inc., USA). FAME were extracted from each isolate in accordance with the MIDI protocol and detected as recommended in the gas chromatography instrument specifications (Agilent 6850; FID, Ultra 2 Column; described above). The acquired FAME profiles of the bacterial isolates were compared with the Aerobic bacterial database (TSBA50) by the Sherlock software system generating similarity indices (IS). Statistics. The results are presented as averages and standard errors over the replicates described above. Bacterial biomass was analyzed using ANOVA and the Tukey post-hoc test.

432


Vol. 50

RESULTS AND DISCUSSION The PLFA based bacterial biomass of the surface layer increased significantly (p < 0.05) in the initial-early stages; it reached maximum in the mid stage and stabilized in the late stage. In contrast, the bacterial biomass of the mineral layer increased throughout the succession, although the increase was not significant until the mid stage (Table II). Bacterial biomass in the mineral layer never caught up with the levels in the surface layer; at all sampling times except for the initial one, bacterial biomass in the mineral layer was significantly lower (50–90 %) than that of the surface layer. This suggests that, after 43 years, in contrast to the surface layer, the process of succession is still continuing in the mineral layer bacterial community. Frouz and Nováková (2004) also found a rapid increase in surface layer microbial biomass during 10–25 years of succession, which peaked in shrub sites and decreased in forest sites. Surface layer microbial biomass, respiration and metabolic activity was the highest in samples from intermediate succession stages (15–25 years) as compared to initial (3- to 14-year-old plots) and older (25–41 years) plots. The later changes in the bacterial population structure occurred at the same time as (i) earthworm-mediated soil mixing which reduced organic carbon content in the surface soil layer and (ii) changes in soil organic matter availability. Table II. Bacterial biomassa of spoil of brown coal colliery substrate Years after deposition

PLFA bacterial biomassb

Proportion of G– bacteriac

Proportion of G+ bacteriac

top layer (0–50 mm) 1 11 21 43

2.99 ± 1.02 12.13 ± 3.60 75.74 ± 14.21 52.90 ± 0.93

29.8 ± 5.5 40.4 ± 7.4 32.1 ± 1.35 39.8 ± 2.8

70.2 ± 59.7 ± 68.0 ± 60.2 ±

9.5 7.8 1.8 7.1

mineral layer (100–150 mm) 1 11 21 43

2.90 ± 4.80 ± 7.59 ± 24.41 ±

0.33 1.80 4.11 9.56

46.0 ± 4.8 37.3 ± 7.9 31.0 ± 1.7 41.0 ± 2.8

54.1 ± 14.8 62.7 ± 8.4 69.0 ± 2.8 59.0 ± 3.6

aMeans ± SE of three replications. bnmol/g dry mass. cIn %; 100 % corresponds to sum of PLFA bacterial biomass estimated according to G+ and G– bacterial

markers.

PLFA biomarkers suggest that G+ rather than G– bacteria dominated the bacterial biomass of our chronosequence samples (Table II). The highest G–/G+ ratio of 0.7 was in early surface samples (11 years old); the values were greater than those in younger (1 year old) or older (21, 43 years old) sites. These findings agree with the species characterization of cultivable bacteria (see below; Table III). The early stage bacterial community contained relatively many fast growing G– proteobacterial species, especially representatives of the Pseudomonas genus. Although these species contributed to the increase of G– bacteria abundance in both layers, the increase of G– biomass was only significant in the surface layer. The most likely reason for the increase in G– bacteria in the surface layer is the presence of roots supplying easily available C (Table I). Schipper et al. (2001) ascribe the changes in microbial hetrotrophic diversity along five plant succession sequences partly to the availability of organic carbon or resources. The ratio of culturable to total cells (C : T) decreased with the age of deposition from the early to the late stage (11–43 years, see Fig. 1). Over 32 years the maximum C : T ratio decreased from 0.039 (±0.005) to 0.016 (±0.0001) in the surface layer, and from 0.068 (±0.020) to 0.028 (±0.004) in the mineral layer, respectively. Maximum C : T ratio was twice higher in the deeper layer of soil than in the upper one. The above data indicated community-level shift resulting in an increasing proportion of nonculturable types of bacteria with time and soil depth. Similar results have been found in a wheat rhizosphere and a continuously stirred tank reactor by Garland et al. (2001) and the presence of an r–K continuum of bacteria in glacial fore field soils was described in the same way (0, 10, 46, 70 and 100 years after deglaciation) (Sigler et al. 2002). Differences were found between the CFCs of bacteria isolated from the surface layer. The early stage (11-year-old) was dominated by fast growers that could quickly colonize the media (75 % of the total bacterial colonies formed in the first 3 d). In samples taken from the mid (21-year-old) and late (43-year-old)

2005


stages, only 30–40 % of the bacterial population could be characterized as fast growers (Fig. 2A). No significant effect of time after deposition in early, mid and late stages was observed in the mineral layer (Fig. 2B). Fontaine et al. (2003) propose that incorporating fresh organic matter (FOM) in a soil may intensify soil organic matter (SOM) mineralization, in what is known as the “priming effect”. They suggest that FOM is used by r-strategists and SOM by K-strategists. In this case the main sources of FOM were herbs, grasses and their roots in the early stage of succession, and woody litter in the older succession sites. A priming effect might thus promote the development of fast growers (r-strategists) at the surface during the early stage. During older successional stages fast growers would be inhibited by the more recalcitrant nature of the FOM and by competition with established slow growing heterotrophic bacteria and fungi.

Fig. 1. Trends in ratio (R) of culturable to total bacteria cell number in primary succession on spoil of brown coal colliery substrate; circles – top layer 0–50 mm, squares – mineral layer 100–150 mm; means ± standard error of 3 replications.

Fig. 2. Colony-forming curves (CFC; see part Culturing in Materials and Methods) for primary succession bacteria inhabiting top layer 0–50 mm (A) and mineral layer 100–150 mm (B); time after deposition of spoil of coal brown colliery substrate: circles – 1 year, squares – 11 years, diamonds – 21 years, triangles – 42 years. CFC were based on the accumulation of 15, 30, 315 and 229 in top layer and 8, 21, 53, and 94 CFU × 106 per g dry soil for mineral layer (average of quadruplicate platings) for 1-, 11-, 21-, and 43-year soils, respectively; means ± SD of 12 replicates.

CFC analysis revealed that the heterotrophic bacterial community on the surface layer of brown coal spoil changed during primary succession from fast growers (r-strategists) toward slow grower bacteria (K-strategists). This shift from r- to K-strategists also occurred more slowly in the mineral layer below. These results support the conclusions of Sigler et al. (2004), who found that early vs. late succession bacterial populations produced by glacial retreat show a similar shift in the proportion of r-strategists. Using a similar approach, we have found a shift from r- to K-strategists in both layers of the soil that developed on brown coal spoil. In our case the initial bacterial population of heterotrophic microorganisms was derived from the pristine microbial community of lacustrine Miocene clay sediments (unpublished data) and any microorganisms that inoculated the spoil sediment during excavation, transport and deposition. These microorganisms are well adapted to (i) using fossil organic carbon as a substrate (the spoil contains no recent organic matter

434


Vol. 50

or plant or invertebrate activity) and (ii) extreme fluctuations of moisture and temperature (Kuráž et al. 2003). In the initial and early stages, PLFA bacterial biomass was low but there was a higher genera and species abundance of cultivable bacteria in comparison with the older succession stages. Cultivating isolates from both layers in the initial stage gave CFCs that were very different from later stages (Fig. 2). Bacteria that produced visible colonies after only 4 d of incubation were dominant (≈60 %) in this soil in both analyzed layers. The high number and species richness of fast growers and high C : T ratio suggest that opportunistic species are present in both layers of the younger successional stages. The incorporation of FOM and SOM changed the intensity of r–K competition in older successional stages, where slow growers prevailed. Species characterization of cultivable bacteria. Four-hundred-seventeen of the 657 isolates analyzed using the MIS Sherlock System could be identified (Table III). The identified isolates were assigned to 35 genera (23 G+ and 12 G–) and 81 species/biotypes (51 G+ and 30 G–). The most abundant genera were Pseudomonas (22 % of total species), Arthrobacter (10 %), Bacillus (9 %) and Paenibacillus (6 %). The Actinobacteria, Proteobacteria and Firmicutes were the main constituents of the bacterial community in all succession stages. The highest genera and species and/or biotype diversity in the initial and early succession stages was on the surface layer. The initial stage was characterized by similar species richness in both layers. In both layers the slowly growing group was more species rich than the fast growing group and was dominated by G+ bacteria. The G+ bacteria continued to dominate the slow grower group during all subsequent stages of succession. G+ bacteria tended to be more ubiquitous than G– species. The greatest shift in the richness of the fast at the expense of slow growers was observed on the early stage in both layers, and was accompanied by important enrichment of fast growers by G– proteobacterial species, especially of genera Pseudomonas. This was also apparent in the mid stage, especially in the mineral layer. The late stage mineral layer also contained a large variety of the fast growing bacteria in contrast to the group of slow growers, but the number of G– species was reduced. This change in the population structure towards fast growing G– bacteria and in particular pseudomonads can be explained by root development and the consequent rhizodeposition of carbon (Killham 1994; Fontaine et al. 2003). The G+ Firmicutes (especially genera Bacillus, Paenibacillus) and Actinobacteria were typical of the late stage in particular in the surface layer. Here, the richness of slow growers counterbalanced the richness of the fast growers. The mid and late stages were characteristic by increased litter input of 1.1 ± 0.4 kg/m3 and 0.5 ± 0.02 kg/m3, respectively, as compared to levels of 0.1 ± 0.03 kg/m3 in the early stage. The litter quality of the study sites differed significantly. The C : N ratio of litter increased with succession age as the wood content increased (22.1 ± 0.4 – early stage, 27.9 ± 1.9 – mid stage, 46.4 ± 3.6 – late stage). These factors doubtless promoted slow growing bacteria (K-strategists) and microfungal litter decomposers. The heterotrophic bacterial population in the surface layer of brown colliery spoil changed with the length of time after deposition. As spoil aged the population changed from r to K growth strategies. In contrast, in the mineral layer the r strategy persisted. We concur with Frouz and Nováková (2004) that the accumulation, transformation and distribution of organic matter is the principal factor controlling the growth strategy of this heterotrophic bacterial population. This research was funded by the Czech Science Foundation grant no. 526/03/1259 and by the Research Plan of the Institute of Soil Biology, Academy of Sciences of the Czech Republic, project no. AV 0Z 6066 0521. We thank Prof. K. Márialigeti (EőtvősLoránd-University, Budapest, Hungary) and to Dr. P. Hill (University of Groningen, The Netherlands) for valuable comments. REFERENCES ANDREWS J.H., HARRIS R.F.: r- and K-selection and microbial ecology, pp. 99–147 in K.C. Marshall (Ed.): Advances in Microbial Ecology, Vol. 9. Plenum Press, New York–London 1986. ATLAS R.M., BARTH R.: Microbial Ecology – Fundamentals and Applications. Benjamin/Cummings Publishing Company Inc., New York 1993. BLOEM J.: Fluorescent staining of microbes for total direct counts, pp. 1–12 in A.D.L. Akkermans, J.D. van Elsas, F.J. de Bruijn (Eds): Molecular Microbial Ecology Manual. Kluwer Academic Publishers, Dordrecht (The Netherlands) 1995. CHIARINI L., BEVIVINO A., DALMASTRI C., NACAMULLI C., TABACCHIONI S.: Influence of plant development, cultivar and soil type on microbial colonization of maize roots. Appl.Soil Ecol. 8, 11–18 (1998). DE LEIJ F.A.A.M., WHIPPS J.M., LYNCH J.M.: The use of colony development for the characterization of bacterial communities in soil and on roots. Microb.Ecol. 27, 81–97 (1993). FONTAINE S., MARIOTTI A., ABBADIE L.: The priming effect of organic matter: a question of microbial competition? Soil Biol.Biochem. 35, 837–843 (2003). FROSTEGÅRD Å., BÅÅTH E.: The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Biol.Fertil. Soils 22, 59–65 (1966). FROUZ J., NOVÁKOVÁ A.: Development of soil microbial properties in topsoil layer during spontaneous succession in heaps after brown coal mining in relation to humus microstructure development. Geoderma 129, 54–64 (2004).

2005


GADGIL M., SOLBRIG O.T.: The concept of r- and K- selection: evidence from wildflowers and some theoretical considerations. Amer. Nat. 106, 14–31 (1972). GARLAND J.L., COOK K.L., ADAMS J.L., KERKHOF L.: Culturability as an indicator of succession in microbial communities. Microb. Ecol. 42, 150–158 (2001). JIANPING SU, YANQING WU, XIAOJUN MA, GAOSEN ZHANG, HUYUAN FENG, YINGHUA ZHANG: Soil microbial counts and identification of culturable bacteria in an extreme by arid zone. Folia Microbiol. 49, 423–429 (2004). KILLHAM K.: Soil Ecology. Cambridge University Press, Cambridge (UK) 1994. KŘÍBEK K., STRNAD M., BOHÁČEK Z., SÝKOROVÁ I., ČEJKA J., SOBALIK Z.: Geochemistry of Miocene lacustrine sediments from the Sokolov Coal Basin (Czech Republic). Internat.J.Coal Geol. 37, 207–233 (1998). KURÁŽ V., MATOUŠEK J., HAJAŠ J., KURÁŽ M.: Preliminary results of the soil physical properties measurement of the spoil heaps under primary succession – brown coal mine area Sokolov, pp. 99–104 in J. Frouz, M. Šourková, J. Frouzová (Eds): Soil Physical Properties and Their Interactions with Spoil Organisms and Roots of Plants. (In Czech with English abstract) Institute of Soil Biology, Acad. Sci. Czech Rep., České Budějovice (Czechia) 2003. MURPHY J., RILEY J.P.: A modified single solution method for the determination of phosphate in natural waters. Anal.Chim.Acta 27, 31–36 (1962). O’LEARY W.M., WILKINSON S.G.: Gram-positive bacteria, pp. 117–201 in C. Ratlege, S.G. Wilkinson (Eds): Microbial Lipids, Vol. 1. Academic Press, London 1988. ORAVECZ O., ELHOTTOVÁ D., KRIŠTŮFEK V., ŠUSTR V., FROUZ J., TŘÍSKA J., MÁRIALIGETI K.: Application of ARDRA and PLFA analysis in characterization of the food, gut and excrement of saprophagous larvae of Penthetria holosericea (Diptera:Bibionidae): a pilot study. Folia Microbiol. 49, 83–93 (2004). SCHIPPER L.A., DEGENS B.P., SPARLING G.P., DUNCAN L.C.: Changes in microbial heterotrophic diversity along five plant successional sequences. Soil Biol.Biochem. 33, 2093–2103 (2001). SIGLER W.V., ZEYER J.: Colony-forming analysis of bacterial community succession in deglaciated soils indicates pioneer stress-tolerant opportunists. Microb.Ecol. 48, 316–323 (2004). SIGLER W.V., CRIVII S., ZEYER J.: Bacterial succession in glacial fore field soils characterized by community structure, activity and opportunistic growth dynamics. Microb.Ecol. 44, 306–316 (2002). SOMMERS L.E., NELSON D.W.: Determination of total phosphorus in soils: a rapid perchloric acid digestion procedure. Proc.Soil Sci. Soc.Am. 29, 902–904 (1972). WATANABE F.S., OLSEN S.R.: Test of ascorbic acid method for determining phosphorus in water and NaHCO3 extracts from soils. Proc.Soil Sci.Soc.Am. 29, 677–680 (1965). WILKINSON S.G.: Gram-negative bacteria, pp. 299–488 in C. Ratlege, S.G. Wilkinson (Eds): Microbial Lipids, Vol. 1. Academic Press, London 1988. ZELLES L., BAI Q.Y., BECK T., BEESE F.: Signature fatty acids in the phospholipids and lipopolysacharide as indicators of microbial biomass and community structure in agricultural soils. Soil Biol.Biochem. 24, 317–323 (1992).