Investigation of the Properties of Sand Tubules, a Biomineralization Product, and their Microbial Community

Abstract

Sand tubules, made up of sand grains cemented by microbe-induced calcium carbonate precipitation, have been found in China's Ningxia Province. Sand tubules grow like a tree's roots about 40-60 cm below the surface. The properties of sand tubules and their bacterial community were examined. X-Ray diffraction analysis revealed that the sand tubules were associated with crystalline calcite. Scanning electron microscopy showed that the crystalline solid had a lamellar structure and lacked the presence of cells, suggesting that no bacteria acted as nucleation sites, nor that the crystalline solid was formed by the aggregation of bacteria. Denaturing gradient gel electrophoresis analysis showed 11 of the 12 detectable bands were uncultured bacteria by BLAST analysis in the GenBank database, and the rest were closely related to Paenibacillus sp. (100% identity). By cultivation techniques, the only strain isolated from the sand tubule was suggested to be related to Paenibacillus sp.; no archaea were found. Furthermore, Paenibacillus sp. was demonstrated to induce calcium carbonate precipitation in vitro.

Introduction

Biomineralization is a process by which an organism forms a mineral. There is a very large diversity of chemical compositions and structures for biominerals, including carbonates, phosphorites, silicates, and Fe- and/or Mn-oxides [8,9,19,20,28]. Biomineralization can be classified as biologically controlled mineralization and biologically induced mineralization [10]. Microbially induced carbonate mineralization has attracted much attention [20]. It plays an important role in a wide range of processes, including geological processes, atmospheric CO2 budgeting, biogeochemical cycling of elements, environmental treatments, conservation of decayed ornamental stone, and civil engineering [2,11,13,25]. Investigations both in nature and in laboratory settings have demonstrated the bacterial precipitation of carbonates [5,15,23]. Some possible mechanisms have been proposed for bacterially induced carbonate minerals precipitation. One of the mechanisms proposed is that urease-producing bacteria, such as Sporosarcina pasteurii, can induce calcium carbonate precipitation by increasing the pH and alkalinity of the environment during the degradation of urea [12,29]. In this proposed mechanism, the physicochemical parameters of the habitats of these microorganisms were altered by their metabolic activities, and allowed the precipitation of minerals to take place. Other researchers have reported that bacteria can serve as a nucleus in the precipitation of Ca ions after adsorption on the cell surface, and that the matrix of extracellular polymeric secretions can affect mineral precipitation on cells [7]. In addition, some bacteria can produce extracellular carbonic anhydrase and induce calcium carbonate precipitation with carbonic anhydrase as the catalyst [16]. From various studies conducted in this field, the phenomenon can be influenced by environmental physicochemical conditions, it is also related both to the metabolic activity and to the cell surface structures of bacteria [24]. Different types of bacteria and abiotic factors (such as salinity and composition of the medium) seem to contribute in a variety of ways to calcium carbonate precipitation in a wide range of different environments.

In the sandy slope of the northwest margin of Zhongwei City, Ningxia Province in the northwestern part of China, some elongated tubes consisting of sand cemented by microbially induced calcium carbonate precipitation (MICCP), tentatively called sand tubules, were found to be present. These structures were hollow and filled with a brown material. Although MICCP is a common phenomenon found in many environments, including sea water, fresh water, industrial waste water, and soil [20], this is a first report on sand tubules. In the present study, the effects of several parameters on calcium carbonate precipitation in sand tubule formation are considered. Among these, ionic composition and concentration of precipitation nuclei are the most important. However, the microbial community is another critical factor. The main goal of the present study was to determine the nature of sand tubules and the bacterial community of the sand tubules by using crystal morphology and molecular biological techniques and traditional cultivation.

 

Materials and Methods

Location Description and Sampling

The location where we discovered the sand tubule is in northwestern China in Zhongwei City near the border of Ningxia and Inner Mongolia. Tengger Desert is close by. The area in which we found the sand tubules is small in size, measuring only about 300 m2. It is on a southern hillside (Fig. 1A) and has a very dry surface with sparse vegetation. On the surface of this area, there were some hollow sand tubule fragments with a dry and hard surface (Fig. 1B). However, about 40-60 cm below the surface, the humid and easily fractured sand tubules were filled with a brown material. The sand tubule was like plant roots extended into the sand (Fig. 1C).

Fig. 1.Sampling location for sand tubules. (A) The sandy slope of sand tubule formation. (B) Fragments of sand tubules on the surface of sandy slope. (C) About 40-60 cm below the surface, the sand tubule grew like roots extending into the sand.

Two sand tubule samples, about 60 cm long each, were removed with sterilized forceps, stored in a sterilized container, and transported immediately to the laboratory at 4℃. One sand tubule was cut into 10 pieces and rinsed with sterile deionized water. Then the brown material in the hollow part of the sand tubule was removed using a sterilized needle and preserved in liquid nitrogen for total DNA extraction. The other tubule was also rinsed with sterile deionized water and disinfected with 0.1 M sodium hypochlorite on its surface, before being cut into 10 pieces. Their terminal end section was submerged into a sterile semisolid medium and cultured at 20℃ for 15 days, aimed at investigating the role of stomata of the sand tubule. The medium contained (g/l) KH2PO4 (0.5), MgSO4·7H2O (0.4), NaCl (0.5), CaCl2·2H2O (0.05), NH4Cl (0.5), peptone (5.0), yeast extract (1.0), FeCl2·4H2O (0.0005), and agar (8.0).

Analysis of Calcium Ion Content

The calcium contents in the sand tubule, the sands surrounding the sand tubule, and ordinary river sand (from the Shangbu River in Hangzhou) were analyzed by atomic absorption spectrometry (AAS). The samples were kept at 105℃ for 24 h before being analyzed. A 1.0 g sample of each was introduced into 200 ml Kjeldahl flasks, digested with aqua regia (12 ml concentrated hydrochloric acid and 4 ml of concentrated nitric acid) at room temperature, and then heated to 95℃. After the fumes had ceased, the mixture was evaporated to dryness on a sand-bath and mixed with 8 ml of aqua regia. Then the mixture was again evaporated until dryness. After evaporation, 10 ml of distilled water was added and mixed with the sample. The resulting mixture was filtered through a blue band filter paper (Whatman). Finally, the filtrate was transferred into a volumetric flask and diluted to 100 ml with distilled water. Each sample was then analyzed for calcium content using an AAS Z-23120 atomic absorption spectrometer (Hitachi, Japan).

Microstructure and Characterization of the Sand Tubule

To observe the microstructure of the sand tubule and to characterize it, pieces were examined under a stereo microscope (Leica M205 FA, Germany) after the brown material had been removed. Then the sand tubule was put into a sample bag and crushed after removal of the inside and surface sand grains. Thereafter, a suspension of the crushed sand tubule was prepared in distilled water using an oscillator. After removal of the sand layer from a piece of sand tubule, the solid crystalline layer was collected. This solid crystalline layer, consisting of small sand grains in a crystalline solid, was wiped manually under a stereo microscope. The crystalline solid was then dried at 50℃ for further analysis. An X'Pert PRO (PANalytical, Netherlands) X-ray diffractometer (XRD) was used to identify the mineral composition of the crystalline solid. A SIRION-100 (FEI, USA) thermal field emission scanning electron microscope (SEM), equipped with a detector for X-ray energy dispersive spectroscopy (EDS) (EDAX GENESIS4000, USA) was used for imaging and elemental analysis of the crystalline solid.

Denaturing Gradient Gel Electrophoresis (DGGE)

DNA from the brown material in the hollow part of the sand tubule was extracted using the UltraClean Soil DNA Isolation Kit (Tiandz Ltd., Beijing, China). Approximately 0.5 g of the brown material (wet weight) was used, and the total DNA leaching liquor was then loaded into the bead column. After a series of washings, the DNA was eluted in 50 μl of TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). The quality of the extracted DNA was checked using 0.8% (w/v) agarose gel electrophoresis. Extractions were quantified using absorption measurements (260 nm) with a UV-2102PCS spectrophotometer (Unico, Shanghai, China), The extracted DNA was stored at -20℃ until further use [6].

The fragments of 16S rRNA genes were amplified from the total DNA using primers F357-GC and R518 [31]. PCR amplification was performed as follows: pre-denaturation at 94℃ for 5 min, denaturation at 94℃ for 0.5 min, touchdown annealing from 65℃ to 55℃ with a decrease of 0.5℃per cycle, followed by 10 cycles at annealing temperature of 55℃, with a final extension of 7 min at 72℃. A 3 μl sample of PCR product was subjected to electrophoresis and analyzed on a UVI gel imaging and analysis system (Syngene Gene Genius, UK).

The PCR products were analyzed by DGGE using a Bio-Rad ᴅ-Code Universal Mutation Detection System (Bio-Rad, Richmond, CA, USA). A n 8% ( w/v) polyacrylamide gel ( acrylamide: bisacrylamide at 37.5:1) was applied to the sample in 0.5 TAE. Optimal separation was achieved with a 30-60% urea formamide denaturing gradient (100% denaturant corresponds to 7 M urea and 40% formamide). Gel runs were carried out for 4 h at 180 V at 60℃ and then gels were stained with 5% Goldview for 30 min. The gel was then photographed and analyzed using the UVI gel imaging and analysis system (Syngene Gene Genius). The representative bands were excised from the DGGE gel and recovered using a gel extraction kit (Sangon Biotech Co., Ltd., Shanghai, China). The recovered DNA was re-amplified with primers F357 and R518, and then sent for sequencing (Sangon Biotech Co.). The sequences were submitted to GenBank (Accession No. KM504262-504273) and identified by comparison with accessible sequences in GenBank with the BLAST server.

Enrichment and Isolation of Microbes

Bacterial enrichments and isolations were carried out using Starkey (STK) medium(for sulfate reducing bacteria), HPBLi and Ren (HPB-LR) medium (for hydrogen-producing bacteria), methanogenium (DSMZ) medium (low salt), minimal medium (MM), and Luria-Bertani (LB) medium. Methanogenium medium (DSMZ) 97 (high salt), methanogenium medium (DSMZ) 823 (moderate salt), hepatocyte medium (HM) (for moderately halophilic bacteria), complete medium (CM) (for Halobacterium cutirubrum) as well as O-S medium (for Halobacterium halobium) were employed in an archaeal search, where 1.8% (w/v) agar was added into the medium to prepare agar plates. For anaerobic enrichments, the medium was dispensed into the serum bottle and flushed with pure N2 for 30 min to purge out dissolved oxygen, and then resazurin was added as the redox indicator (the final resazurin concentration was 1.0 μg/ml). The serum bottle was sealed with a rubber stopper and aluminum, and then autoclaved at 120℃ for 20 min [33]. The material in the hollow part of the sand tubule was collected in the anaerobic chamber, and then 0.1 g of the sample was mixed with 0.9ml of the medium. After enrichment by corresponding culture media at 18℃ (the mean temperature of sampling site) for 48 h, the culture was serially diluted with the same medium and repeated plating onto agar plates until single colonies were obtained. Single colonies were transferred several times. The isolates were identified by analyzing the partial sequence of the gene encoding 16S rRNA. The bacterial genomic DNA was extracted using a commercially available extraction kit (Sangon Biotech Co.). The PCR amplification of the 16S rRNA gene was carried out using bacteria-specific universal primers 7f (5'-CAGAGTTTGATCCTGGCT-3')/1540r (5'-AGGAGGTGATCCAGCCGCA-3') by a thermal cycler. The PCR products were purified and sequenced commercially. The resultant sequence was submitted to GenBank.

Calcium Carbonate Precipitation Experiments

In the study, calcium carbonate precipitation experiments were carried out in calcium chloride liquid medium composed of 5.0 g/l yeast extract, 10.0 g/l protease peptone, 5.0 g/l NaCl, and 10 g/l calcium chloride, and the pH was adjusted to 7.2 with 0.1 M NaOH. The culture media were autoclaved at 112℃ for 20 min. The Paenibacillus sp. strain was cultivated under aerobic conditions at 18℃ for 48 h in LB medium. Then 0.2% of the seed culture (cellular density = 106 cells/ml) was transferred into calcium chloride liquid medium and incubated at 18℃/180 rpm on a horizontal shaker (Sangon Biotech Co.). After incubating for 120 h, crystals were collected from the medium, transferred to distilled water, and washed to remove impurities. The purified crystals were then air dried at 50℃ and examined by SEM and XRD [14]. To eliminate the effects of culture media and seed medium on the biomineralization, a control consisting of uninoculated culture media and media inoculated with autoclaved seed culture was included in all experiments.

 

Results and Discussion

Microstructure and Characterization of the Sand Tubules

Whereas the terminal end of the sand tubule was comparatively thick, having a diameter of more than 2.6 cm, it had a very narrow head end, with a diameter of only 0.8 cm. Many small holes were found at the surface near the head end of the sand tubule (Fig. 2), but there were a few small holes at the surface closer to the terminal end. On observing the sections, we found that the sand grains were cemented together, forming the crystalline solid of the sand tubule (Fig. 3A). Other sand grains around the sand tubule were dispersed because of a lack of crystalline solid (Fig. 3B). In addition, there were some little bubbles (about 0.5 mm in diameter) appearing at the sand tubule holes after being immersed in culture for 6 days in a semisolid medium. However, there was no similar phenomenon showing up on the sand tubule treated at 121℃ for 10 min. We believe that the holes in the sand tubule were blowholes that served as channels for gas exchange by the sand tubule microbes. The formation and distribution of the blowholes could be related to microbial metabolism. The greater number of blowholes at the terminal end of the sand tubule may be conducive to faster growth of microbes and extension of the sand tubule.

Fig. 2.Stomas on the surface at the end of the sand tubule. (A) Overview. (B) Stoma (mm, millimeter; dmm, decimillimeter).

Fig. 3.Images of the sand tubule profile and sands around the sand tubule. (A) Section of sand tubule; sands were cemented together to form a sand tubule by crystalline solid. (B) Other sands around the sand tubule were dispersed individually (dmm, decimillimeter).

Analysis of Calcium Ion Content

According to earlier research, the concentration and varieties of calcium ion are the most important in influencing MICCP [19,23,24]. To understand the role of calcium in sand tubule formation, we determined the calcium contents of the sand surrounding the sand tubule, of the sand tubule, and of ordinary river sand, AAS. The respective calcium contents were 12.52, 46.71, and 1.76 mg/g (Fig. 4), showing that the calcium content of the sand tubule and the surrounding sand was much higher than that of ordinary river sand. The higher calcium salt content in the sand tubule may be the material basis of the sand tubule. In addition, the calcium content in the sand tubule was over 3.7 times that of the surrounding sand, suggesting that the microbes in the sand tubule can enrich the calcium from the surrounding sand and form a calcium deposit by agglomeration of the sand.

Fig. 4.The calcium ion content of different samples. (1) The sands surrounding sand tubule; (2) the sand tubule; and (3) ordinary river sand.

Microstructure and Characterization of Crystalline Solids in Sand Tubules

SEM. The morphological feature of the crystalline solids in the sand tubule was examined under SEM (Fig. 5). Only lamellar shape was observed in the crystalline solid. No other typical shapes, such as spheres, hemispheres, dumbbell shapes, or rhombohedra were seen or reported by other authors [15,26,32]. The morphology of bacterial carbonate precipitates is quite heterogeneous, and it can be influenced by microorganisms, environmental conditions, and enzyme activity [22]. Crystal growth can also be inhibited or altered in the presence of proteins, organic compounds, or inorganic compounds, which can inhibit calcium carbonate precipitation by adsorption to planes of growing crystals [21]. Moreover, different calcium salts can induce different crystal sizes and crystal morphology [14]. Many researchers have reported that calcium carbonate precipitation produced by the bacteria showed a significant quantity of cell marks inside the precipitate, as well as on the surface [12,18,29], which suggests that the bacteria served as nucleation sites for calcium carbonate precipitation upon adsorbing calcium ions onto the cell surface during the mineralization process. However, there were no cell marks in the crystalline solid of the sand tubule (Fig. 5), suggesting that the crystalline solid was not formed by bacterial aggregation. Thus, in the case of sand tubules, the microorganisms contributed to their formation by changing components in the surrounding sand by their metabolism. This confirms our previous hypothesis that mineralizing microorganisms only exist inside the sand tubule and protect the tubules by cementing sand grains.

Fig. 5.Scanning electron micrographs of crystalline solid from the sand tubule. (A) General overview of the crystalline solid. (B) Details of the crystalline solid.

EDX microanalyses of isolated crystalline solid from the sand tubule are presented in Fig. 6. The composition of the crystalline solid is very complicated. It has variable amounts of Si, Ca, and Al, but also has a low content of K, Fe, Mg, Na, and Ti. Moreover, P is absent in the crystalline solid. This indicates that some crystalline solids do not contain calcium phosphate, which commonly occurs in one of two common microbially induced calcium biomineralization products. Moreover, phosphate actually inhibits calcium carbonate precipitation [3,4].

Fig. 6.EDX spectra of crystalline solid from the sand tubule.

XRD. On the basis of the relative intensities of the diffraction peaks, it can be inferred that the most abundant component of the crystalline solid in the sand tubule is quartz (Fig. 7), the main component of sand grains. It is noteworthy that CaCO3 crystal is identified as calcite, not aragonite or vaterite, which are more common in CaCO3 precipitate induced by microorganisms [25]. In addition, there are a number of illites and kaolinites in the crystalline solid (Fig. 7). As they are common ingredients in the sand of the sampling site, we surmise that they may come from the surrounding sand grains rather than being induced by microorganism.

Fig. 7.XRD profiles of crystalline solid from the sand tubule.

Fingerprinting of Bacterial Community by PCR-DGGE

No microbes were found in the crystalline solids in sand tubules, so bacterial community investigation was made from the inclusion (brown material) in the sand tubule. In this work, about 250 bp 16S rRNA gene V3 region PCR products were analyzed by DGGE to reveal the bacterial composition and diversity in the sand tubule. As shown in Fig. 8, there were 12 detectable bands in the DGGE profiles. Among these bands, Nos. 1, 2, 4, 5, and 6 were dominant. In order to identify the bacterial species, the 12 bands in the DGGE profiles were eluted from the polyacrylamide gels and sequenced after re-amplification; the corresponding identifications are reported in Table 1. Although there are no exact 16S rRNA gene similarity limits for defining specific taxa, in general, species definition requires sequence similarities greater than 98% [30]. Only band 4 showed 100% similarity to 16S rRNA gene sequences from GenBank databases and was identified as representing Paenibacillus (Table 1). The other 11 bands were assigned to uncultured bacteria, and their classification had not yet been described in GenBank databases. The sequence similarity of bands 1, 2, and 6 to sequences in the GenBank database was 95%, 95%, and 96%, respectively. The low sequence similarities indicated that they belong to a new and not yet named phylogenetic group.

Fig. 8.PCR-DGGE analysis of uncultivable bacteria within the sand tubule. The directly extracted DNA from the sand tubule was used as the template for PCR amplification with 357F-GC/518R. The obtained PCR products were separated by DGGE. The bands in the gel were recovered by PCR and identified by sequence analysis.

Table 1.Only highest homology matches are presented. aBands are numbered according to Fig. 8. bIdentity represents the % identity shared with the sequences in the GenBank databases. CPutative phylum was obtained by comparison in the Ribosomal Database Project (RDP, http://rdp.cme.msu.edu/seqmatch/seqmatch_intro.jsp).

Isolation of Microbes by Cultivation Techniques

To enhance our insight into the microbial communities of the sand tubule, both bacteria and archaea were enriched and isolated using a series of culturing media. Only one bacterium (but no archaea or anaerobic bacteria) was obtained in the limited conditions of our assessment. The 16S rRNA gene fragment of the strain was amplified and sequenced, and then submitted to and compared with GenBank databases (Accession No. KF669895). The result showed that the affiliation of the strain was Paenibacillus sp. (100% identity). We noticed that the strain Paenibacillus sp. was also detected by PCR-DGGE (Table 1). These suggested that molecular biological techniques and traditional cultivation might complement and validate each other in analyzing the microbial community of the sand tubule. Therefore, we believed that Paenibacillus sp. must be one of the bacteria in the sand tubule. Many previous reports have associated MICCP with Sporosarcina pasteurii, B. megaterium, Kocuria flava, B. sphaericus, B. pseudofirmus, B. subtilis, Myxococcus xanthus, B. halodurans, B. cohnii, and B. pseudofirmus [1,17,18,20]. However, Paenibacillus sp. associated with calcium carbonate precipitation has seldom been reported in previous publications [27].

Characterization of Precipitate Crystals in In Vitro Environments

To investigate the ability of MICCP by Paenibacillus sp. isolated from the sand tubule, CaCO3 precipitation experiments in in vitro environments were carried out. The film on the conical flask wall and precipitate crystals were observed after culture for 72 h on a horizontal shaker. None was detected in uninoculated control media or media inoculated with dead bacteria. The morphology and chemical constituents of the precipitate crystals were analyzed with SEM and XRD. SEM analysis of the microbial precipitate formed in artificial media revealed the presence of both rhombohedra as well as spherical crystals (Fig. 9). They have great difference with the lamellar shape of the sand tubule. However it was the same as the crystalline solid of the sand tubule in that no cell mark appeared in precipitated crystals in artificial media, also suggesting that the bacteria do not serve as the nucleation loci. XRD revealed the presence of both calcite and vaterite as the major components present in microbial precipitates formed in artificial media (Fig. 10), whereas only calcite was observed as a predominant component in the crystalline solid of the sand tubule.

Fig. 9.Scanning electron micrographs of precipitate crystals in artificial media. (A) General overview of crystalline solids. (B) Details of crystalline solids.

Fig. 10.XRD profiles of precipitate crystals in artificial media.

Paenibacillus sp., isolated from the sand tubule, has the ability of MICCP. It is entirely possible that it plays a very important role in contributing to the formation of the sand tubules. Although the morphologies and chemical constituents of the crystals do not completely match in play in the in situ versus in vitro environments, they are quite heterogeneous and can be influenced by environmental conditions [14,21,22]. In addition, vaterite is unstable owing to its higher solubility and lower density as compared with calcite. In aqueous solution, it rapidly transforms into the latter phases [25]. It is also worth remembering that no cell marks appeared in the precipitated crystals from artificial cultures and sand tubules, and that no bacteria served as nucleation sites. Moreover, it is possible that the microorganisms precipitating the sand tubules in situ are not culturable in the limited conditions assessed without involving Paenibacillus sp. Therefore, further work needs to be done to see how these microorganisms contribute to the formation of the sand tubules in situ.