JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 17 July 2007
Published in J Environ Qual 36:1249-1259 (2007)
DOI: 10.2134/jeq2006.0315
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Agricola
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Related Collections
Right arrow Ecosystem Restoration
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals

TECHNICAL REPORTS

Bioremediation and Biodegradation

Bacterial Community Changes during Plant Establishment at the San Pedro River Mine Tailings Site

Karyna Rosarioa, Sadie L. Iversona, David A. Hendersonb, Shawna Chartranda, Casey McKeonc, Edward P. Glennc and Raina M. Maiera,*

a Dep. of Soil, Water, and Environmental Science, Univ. of Arizona, Tucson, AZ 85721-0038
b Dep. of Animal Sciences, Univ. of Arizona, Tucson, AZ 85721-0038
c Environmental Research Lab, 2601 E. Airport Drive, Tucson, AZ 85706

* Corresponding author (rmaier{at}ag.arizona.edu).

Received for publication August 7, 2006.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
Mine tailings are moderately to severely impacted sites that lack normal plant cover, soil structure and development, and the associated microbial community. In arid and semiarid environments, tailings and their associated contaminants are prone to eolian dispersion and water erosion, thus becoming sources of metal contamination. One approach to minimize or eliminate these processes is to establish a permanent vegetation cover on tailings piles. Here we report a revegetation trial conducted at a moderately impacted mine tailings site in southern Arizona. A salt and drought-tolerant plant, four-wing saltbush [Atriplex canescens (Pursh) Nutt.], was chosen for the trial. A series of 3 by 3 m plots were established in quadruplicate on the test site to evaluate growth of four-wing saltbush transplants alone or with compost addition. Results show that >80% of the transplanted saltbush survived after 1.5 yr in both treatments. Enumeration of heterotrophs and community structure analysis were conducted to monitor bacterial community changes during plant establishment as an indicator of plant and soil health. The bacterial community was evaluated using denaturing gradient gel electrophoresis (DGGE) analysis of 16S rDNA PCR gene products from tailings samples taken beneath transplant canopies. Significant differences in heterotrophic counts and community composition were observed between the two treatments and unplanted controls throughout the trial, but treatment effects were not observed. The results suggest that compost is not necessary for plant establishment at this site and that plants, rather than added compost, is the primary factor enhancing bacterial heterotrophic counts and affecting community composition.

Abbreviations: ANOVA, analysis of variance • BLM, Bureau of Land Management • C, plots containing transplants with compost • CECL, University of Arizona Chorover Enviromental Chemistry Laboratory • CFU, colony-forming unit • DGGE, denaturing gradient gel electrophoresis • HIC, University of Arizona Superfund Basic Research Program Hazard Identification Core • ICP-MS, inductively coupled plasma–mass spectrometry • KNMDS, Kruskal's Non-Metric Multidimensional Scaling • NC, plots containing transplants with no compost, OSC, off-site control plots • SPRNCA, San Pedro River National Conservation Area • U, unplanted control plots • WQCL, University of Arizona Water Quality Center Lab


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
THERE is emerging concern regarding the environmental consequences of mine tailings sites, especially those in sensitive arid and semiarid riparian or wildlife refuge areas. Tailings are produced during ore processing and are characterized by little or no vegetation, poor soil structure, low nutrient content, surface crusting, and high levels of contaminants including heavy metals (Ledin and Pedersen, 1996). The impact of tailings goes beyond the immediate site because erosion and leaching processes release and spread tailings contaminants to other areas. Removal of tailings materials is costly largely because they are extensive in nature and considered to be hazardous waste. Thus, remediation efforts have focused on stabilization of the tailings.

One approach to stabilization is phytoremediation or establishment of a plant cover that acts to reduce all spreading mechanisms: eolian dispersion, water erosion, and leaching. Concomitantly, a plant cover acts to stabilize soil contaminants by altering the water flux through the soil, by incorporating free contaminants into the roots and rhizosphere, and by improving ecosystem structure and function within the tailings (Cunningham et al., 1995). The challenge with phytostabilization in arid and semiarid environments is that plant establishment is often limited by high levels of metals and extreme pH in addition to the other challenges that arid environments pose to plants (i.e., low water availability, low organic matter, salinity, and high temperatures). In general, the development of a persistent vegetative cover depends on selecting plant species suitable to site conditions and amending tailings with organic matter that improves soil properties that favor plant growth.

The improvements made to mine tailings sites through phytostabilization rely, in part, on activities of soil microorganisms that are essential to healthy ecosystems (Crowley et al., 1997). Bacteria play an important role in the success of phytostabilization by promoting plant growth through the mineralization of inorganic nutrients (Xu and Johnson, 1995), production of growth regulatory substances like phytohormones (Pishchik et al., 2002), facilitation of nutrient uptake from the soil (Burd et al., 2000), improvement of soil structure by soil aggregation (Maier et al., 2000), and reduction of metal toxicity (Salt et al., 1999; Sandrin et al., 2000). In return, established plants provide a nutrient-rich environment that can stimulate bacterial activity (Glick, 2003; Kuiper et al., 2004). Thus, monitoring changes in the microbial community can aid in assessing the success of reclamation strategies.

The goal of this study was to use the local native shrub four-wing saltbush to revegetate barren areas in a former silver-gold mill site located adjacent to the San Pedro River in southern Arizona and determine if plant establishment had a measurable influence on the bacterial community. The site is in the San Pedro riparian area, which was designated as a National Conservation Area in 1988 to help protect the fast disappearing desert riparian ecosystem in the Southwest. A revegetation trial was conducted using four-wing saltbush transplants to investigate if this species was suitable for phytostabilization purposes. The specific objectives of the revegetation trial were (i) to determine if four-wing saltbush could be successfully established in the tailings, (ii) to evaluate if the addition of compost amendment in the tailings was necessary to achieve this goal, and (iii) to follow changes in the bacterial community during the revegetation process. Bacterial community responses beneath transplant canopies were studied by monitoring changes in the abundance of heterotrophic bacteria (culturable counts) and in the bacterial community structure based on DGGE community fingerprints using universal bacterial primers.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 Conclusions
 REFERENCES
 
Site Background and Study Area Description
The Boston mill mine tailings site (Fig. 1) is adjacent to the San Pedro River in the San Pedro River National Conservation Area (SPRNCA), approximately 3.2 km (2 miles) south of Fairbank, AZ, and 12.6 km (8 miles) southwest of Tombstone, AZ, at an elevation of 1192 m (31°40'38.6''N, 10°10'53.4''W). It is one of five mill sites along the upper San Pedro River, within the SPRNCA, that were active from 1879 to 1887 for processing silver and gold ore that was brought from mines in Tombstone, AZ. The site covers approximately 2.8 ha (7 acres) and can be divided into three areas; an upper tier near the mill foundation, a lower tier at a slightly lower elevation south of the upper tier, and tailings across the railroad tracks in the floodplain. Preliminary data provided by the Bureau of Land Management (BLM), which currently manages the site indicated that the upper tier has the highest concentrations of heavy metals. This area has a shallow layer of tailings (10–20 cm) and includes completely barren areas to the west of the mill foundation and some areas with patchy vegetation on the eastern side composed primarily of big sacaton grass (Sporobolus wrighti Munro ex Scribn) and mesquite (Prosopis spp.). This partial vegetation suggests that natural attenuation has slowly occurred on the tailings during the past century. For this reason the Boston mill site is considered to be a moderately impacted mine tailings site. Nevertheless, barren areas still pose a source of concern for eolian dispersion and water erosion of tailings into the San Pedro River and the surrounding region, which is a wildlife refuge and hiking area.


Figure 1
View larger version (93K):
[in this window]
[in a new window]

  		Fig. 1. Photos of the study area (A) showing an aerial view of the site and the relative locations of the transplant plot area, the control site, and their proximity to the San Pedro River. The line bisecting the site from the lower left of the photo (northwest) to the mid-right (southeast) is a railroad track. Photo (B) shows a closeup of two of the 3 by 3 m plots just after having received the transplants. The inset photo is an example of a transplant.

 
Experimental Design and Sampling Summary
A 24 by 30 m study area in the upper tier of the Boston Mill site was selected for the revegetation trial (Fig. 1). One of the objectives of this work was to determine whether compost amendment was necessary for establishment of four-wing saltbush on the site. Therefore, two treatments were evaluated: four-wing saltbush transplants with compost amendment (C), and unamended transplants (NC). A 6 by 12 m off-site control (OSC) area (no mine tailings) was chosen southeast of the study site. For each treatment (compost, C; no compost, NC; and off-site control, OSC) quadruplicate plots (n = 4) were established on the relevant study area following a complete randomized design. Each plot received five transplants and plant growth was evaluated by height and crown measurements made on each plant at time 0, 11, and 18 mo (n = 20). A soil sample was taken from each plot for chemical, physical, and microbiological analysis before planting (n = 4). Following planting, root zone soil samples were taken from one plant in each plot at 3, 4, 5, 7, 8.5, 11, 13, 16, and 18 mo for microbial analysis (n = 4). Shoot tissue samples were collected at 18 mo to determine plant shoot metal content. Tissue from each of the five plants in a plot was harvested and combined into a single composite sample for each NC and C treatment plot (n = 4). It should be noted that transplants did not survive in the OSC area, and so for shoot metal content comparison four composite samples were taken from nearby indigenous four-wing saltbush shrubs to obtain background shoot tissue metal levels (n = 4).

For the microbial analysis a third treatment, unplanted control (U), was added to account for natural variations in the bacterial community. The U samples were initially taken from two areas (n = 2). The first was an area 1.7 m outside the west side of the 24 by 30 m trial area and the second was just inside the 24 by 30 m trial area on the west side. Thirteen months following planting, two more U sites were randomly selected on the east side of the plot and monitored from 13 to 18 mo (n = 4).

Transplants
The plant chosen for this study was four-wing saltbush, which is a perennial halophytic subshrub native to Arizona, California, Nevada, and Utah (USDA-NRCS 2005). This plant is considered drought tolerant and has been observed encroaching into other historical mine tailings sites (USDA-SCS, 1977; Booth et al., 1999; Arunachalam et al., 2004; Jefferson, 2004) as well as the Boston Mill site. Four-wing saltbush transplants were germinated and grown in the greenhouse for 2 mo in a 2:1 sand/compost mix in 0.375-L pots. The compost was a commercial composted mulch (Sunshine mix no. 1, Sun Gro Horticulture). This material was also used as the compost amendment in the field trial (Tables 1 and 2).


View this table:
[in this window]
[in a new window]

  		Table 1. Selected characteristics for treatment plot tailings samples, the off-site control soil, and the compost used in this study.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Total metal concentrations (mg kg–1) for the treatment plot tailings, the off-site control soil, and the compost used in this study.

 
Field Study
Planting took place on 31 Oct. 2003. The study area was divided into eight 7.5 by 12 m plots. A point within each plot was selected using random X and Y coordinates and marked with a pin flag to serve as a reference to later establish the 3 by 3 m treatment areas in each plot (Fig. 1). Quadruplicate plots per treatment were established on the study area following a complete randomized design. To establish the treatments on the field, a 3 by 3 m quadrat was centered on a random point that was previously marked with a pin flag in each of the eight 7.5 by 12 m plots. The four corners of the quadrat were flagged and one transplant was placed near the inside of each of the quadrat corners and one in the middle. The transplants were planted by digging a hole deep enough to cover the roots, centering the transplant in the hole, and filling it back with the original soil (Treatment NC) or with approximately 3.8 L (1 gallon) of compost (Treatment C). For Treatment C, half a gallon of compost was added first, followed by a thin layer of soil, the other half of the compost, and finally a thin layer of soil at the top. The 6 by 12 m OSC area similarly received transplants in four 3 by 3 m plots that were evenly spaced in the area. After planting, all transplants were covered with chicken wire cages for approximately 6 mo to protect against foraging animals. Over the next 8 mo the transplants were irrigated with a total of 44 L of water per plant using the following schedule: Day 0 and 3, 1.5 L; Day 7, 4 L; Day 14, 3 L; Day 24, 4 L; Days 40, 45, 52, 3 L; Days 83, 100, 4 L; Days 114, 131, 181, 3 L; and on Day 218, 4 L.

The height (h) and crown diameters (d1 and d2) of the transplants were measured immediately after planting and at 11 and 18 mo to assess growth. The height measurement was made from the ground to the highest point on the shrub. The two crown diameter measurements were made perpendicular to each other at the widest part of the shrub canopy. Shrub volume was estimated by assuming an elliptical shrub shape (V = 4/3{pi}(h/2 x d1/2 x d2/2) (Thomson et al., 1998).

Soil and Plant Sampling
Soil samples were taken from each of the C and NC treatment plots (n = 4) before planting and from the U plots (n = 4) at the end of the trial to determine selected physical and chemical characteristics. A 500-g surface sample (5–10 cm) was taken from each plot using a closed bucket auger and stored in a plastic bag for analysis.

For microbiological analysis, 5-g soil samples were taken (5–10 cm) using a sterile spatula (washed with 95% ethanol and flamed between samples), placed into sterile vials, and stored at 4°C within 4 h of collection for microbial analysis. These samples were collected at 0, 3, 4, 5, 7, 8.5, 11, 13, 16, and 18 mo to follow bacterial community changes during plant establishment. The time zero samples were taken before planting as described above. Following planting, one plant was chosen from each treatment plot (NC and C treatments, n = 4) for the microbial analysis. Selected plants looked healthy, and with the exception of plot C3, were at least 0.5 m away from any encroaching vegetation. Soil samples were taken beneath the transplant canopy of the chosen plant within 2 cm of the stem and approximately 5 cm deep. For the unplanted controls, two U plots were sampled from 3 to 11 mo (n = 2) and two more U plots were added from 13 to 18 mo (n = 4). U plots samples were taken within 5 cm of the pin flag marking a random point.

Shoot tissue samples were collected 18 mo after planting from each NC and C treatment plot to determine plant shoot metal content. Since transplants did not survive in the off-site control area, four composite samples were taken from nearby indigenous four-wing saltbush shrubs to obtain background shoot tissue metal levels.

Soil and Plant Analysis
Soil physical and chemical properties were determined by the Water Quality Center Lab (WQCL) and by the Chorover Enviromental Chemistry Laboratory (CECL) of the University of Arizona using standard soil analysis methods (Tables 1 and 2). Total metal concentrations were determined following microwave-assisted digestion of dry soil samples (USEPA Method 3051) and analysis by inductively coupled plasma–mass spectrometry (ICP-MS).

For plant tissue total metal content, samples were carefully rinsed with nanopure water to remove soil particles, dried at 65°C for 2 d, and digested following USEPA Method 3051. Samples were analyzed using ICP-MS by the University of Arizona Superfund Basic Research Program Hazard Identification Core (HIC).

Heterotrophic Bacterial Counts
Heterotrophic counts were performed using R2A agar medium (Becton, Dickinson and Co., Sparks, MD) with 200 mg L–1 of cyclohexamide to prevent fungal growth. Each sample was plated in triplicate. Briefly, inocula were prepared by mixing 1 g (wet wt.) of soil with 9.5 mL sterile Zwittergent (CalBiochem, La Jolla, CA) extractant (per L H2O: 2 g NaCl, 0.0002% Zwittergent). The sample was vortexed for 2 min and then serially diluted in sterile water, plated on R2A, incubated at 23°C in the dark, and enumerated after 4 d.

Soil Bacterial Community DNA Extraction and Purification
DNA was extracted from soil bacterial communities by direct lysis using the Fast DNA Spin Kit for Soil (Qbiogene, Carlsbad, CA). The manufacturer's protocol was followed with some modifications. Before the extraction all sterile tubes, caps, and solutions used in the protocol were exposed to UV light for 5 min in a UVC-508 Ultraviolet Crosslinker (Ultra-Lum, Claremont, CA) to remove any potential DNA contamination. The bead beating procedure was performed by attaching the lysing matrix tubes to a Vortex Genie II vortexer (VWR, West Chester, PA) using a Mo Bio adaptor (Mo Bio Laboratories, Solana Beach, CA) and vortexing the tubes at maximum speed for 10 min. Cell debris, soil particles, and the lysing matrix beads were separated from the DNA-containing supernatant by centrifugation at 14000 x g for 15 min. An alternative manufacturer's protocol was followed to enhance removal of soil humic materials that could inhibit PCR amplification. Samples with supernatants retaining a brown color were assumed to contain humic material. For these samples the Binding Matrix-DNA complex was rinsed repeatedly with saturated 5.5 M guanidine thiocyanate (GTC, Sigma, St. Louis, MO) until the supernatant lost its brown tint.

16S rDNA PCR for Denaturing Gradient Gel Electrophoresis
A 391 bp (including the GC clamp) portion of the 16S rRNA gene (rDNA) encompassing the hypervariable V9 region of the domain Bacteria was amplified to target the soil bacterial community in DNA extracts using primers 1070f and 1406r-GC (Ferris et al., 1996). The PCR for DGGE was accomplished using a modified protocol from Colores et al. (2000). Briefly, 50 µL reactions contained 1X Expand Hi Fidelity PCR Buffer with 1.5 mM MgCl2 (Roche, Indianapolis, IN), 0.5 µM of each primer, 0.4 g L–1 unacetylated bovine serum albumin (Sigma, St. Louis, MO) to relieve PCR inhibition (Kreader, 1996), 200 µM of each deoxynucleoside triphosphate, 5% dimethyl sulfoxide, 1.4 U Expand High Fidelity PCR System Enzyme (Roche, Indianapolis, IN), and 0.2 to 0.28 mg L–1 soil bacterial DNA extract. The reactions were exposed to 30 cycles at 95°C for 45 s, 55°C for 45 s, and 72°C for 30 s in a GeneAmp 2400 PCR System thermal cycler (PE Applied Biosystems, Foster City, CA).

The PCR products (5 µL per lane) were visualized and evaluated using electrophoresis through a 2% GenePure LE agarose gel (Intermountain Scientific Corp., Kaysville, UT) followed by ethidium bromide staining. Products in lanes that showed two bands, the desired product and another of slightly larger size, were rejected because DGGE analysis indicated that the extra band was associated with template overloading. Rejected extracts were reamplified using a reduced template concentration. Most of the soil DNA extracts produced single bands using an approximate concentration of 0.28 mg L–1, but those requiring reamplification were diluted incrementally down to 0.2 mg L–1 until only a single band was observed.

Denaturing Gradient Gel Electrophoresis
The DGGE analysis was performed using a Dcode Universal Mutation Detection System (Bio-Rad Laboratories, Hercules, CA). Acrylamide gels (6%) were prepared with a 50 to 60% urea–formamide denaturing gradient according to the manufacturer's protocol. Gel lanes were loaded with 25 to 35 µL of PCR product depending on its intensity on the agarose gel. The gels were run at a constant voltage of 50 V for 19 h at 60°C, stained with 3X SYBR Green I (Molecular Probes, Eugene, OR) for 40 min, and then imaged. To compensate for inconsistencies in electrophoresis, the gels were aligned by loading an external reference ladder at regular intervals. The ladder was composed of pooled PCR product from isolated heterotrophic bacteria from tailings samples. The DGGE profiles (banding patterns) were manually scored according to the method of Konopka et al. (1999) and evaluated using Quantity One 4.5.2 software (Bio-Rad Laboratories, Hercules, CA).

Statistical Analysis
All statistics were performed using the statistical software package R (R Foundation for Statistical Computing, Vienna, Austria). Data were all found to be normally distributed with constant variance. Results of all F tests were considered significant at 95% confidence level. For the revegetation trial, data were subjected to a two-way analysis of variance (ANOVA) for a completely randomized design, with sample date (11 or 18 mo) and soil treatment (compost or no compost) as independent variables and plant volume as the dependent variable. Dead plants were excluded from the data analysis. The ANOVA had 1 df due to sample date, 1 df due to soil treatment, and 64 df due to error. Systat 11.0 software (Systat Software, San Jose, CA) was used to conduct the analysis of variance. Plant volumes were log transformed to equalize variances among treatments. The hypotheses of unequal variance between soil treatments or sample dates were not significant (p > 0.05) for transformed data. At the end of the trial, differences in shoot metal concentrations were analyzed using one way ANOVA of a completely randomized design. Pearson's correlation coefficient (r) was used to screen linear relationships between plant and soil metal concentrations. For the bacterial analysis, treatment effects on heterotrophic culturable counts as a function of time were analyzed using a nested ANOVA model with a maximum likelihood estimator on log transformed data. Treatment was set as a fixed factor, while plant and time effects were set as random factors. Treatment differences were evaluated using pairwise t tests on weighted means calculated for each treatment level (i.e., least squares means, {alpha} = 0.05).

The profiles obtained from DGGE gels were evaluated in three ways. First, ANOVA analysis of the number of bands in each profile was used to evaluate whether there were treatment effects on species richness (the number of populations present). Second, Dice coefficient analysis was used to evaluate the similarities between individual profiles within each treatment plot (C1, C2, C3, C4, NC1, NC2, NC3, NC4, U1, and U2) as a function of time. This analysis examined the extent of the community structure change that occurred in each plot between consecutive sampling times during the field trial. The Dice coefficient is expressed as S = 2a/(b + c) where a represents the number of bands present in both profiles, b refers to the number of bands present in Profile 1, and c represents the number of bands present in Profile 2. Finally, Kruskal's Non-Metric Multidimensional Scaling (KNMDS; Venables and Ripley, 2002) was used to evaluate similarities among profiles from all treatments at each time point. This nonparametric ordination method was used to visualize and interpret changes in the bacterial community during the revegetation trial based on binary distance where similar bacterial communities (i.e., profiles) cluster more closely than those with low similarity in multidimensional space. Differences between bacterial communities were evaluated with a permutation test ({alpha} = 0.05). A stress factor (sf) was also calculated to reflect goodness-of-fit of the model (sf < 0.1 was considered a good fit). The configurations in multidimensional space were evaluated using three or four dimensions (D) to minimize stress. Comparisons of the bacterial community could be performed only for samples within the same gel. Therefore, treatment and time effects were analyzed individually due to a limited number of wells per gel.


   Results
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
Discussion
 Conclusions
 REFERENCES
 
Site Characterization
Selected physical and chemical characteristics were determined for samples from each treatment plot: NC, C, U, and OSC (Table 1). The alkaline pH and the presence of organic carbon are key parameters that indicate this as a moderately stressed site. In severely stressed sites, the pH is usually acidic (<3 to 4) and there is essentially no organic matter. In this site the pH is moderately alkaline, ranging from 8.3 to as high as 9.8 in one sample and total organic carbon values ranged from 8.6 to 21.4 mg kg–1 with an average of 11.6 mg kg–1. These values can be compared very favorably to the OSC, which has a pH of 8.7 and a total organic C content of 4.8 mg kg–1. The big difference between the tailings and the OSC is the total metal concentrations, which were greater in the mine tailings samples than the OSC sample for all metals measured except Al (Table 2). In particular, Pb, Hg, and Cd were elevated up to 100-fold over the OSC.

Revegetation Trial
One goal of this 18-mo revegetation field trial was to evaluate whether four-wing saltbush transplants would survive and grow in a moderately impacted mine tailings site and if so, whether compost addition was necessary. Results from this study show that after 1.5 yr, more than 80% of the transplants survived in the tailings area regardless of whether or not compost was added during planting. A significant increase in growth (p < 0.05) was observed during the first 11 mo of the study with average plant volumes increasing from 0.00047 ± 0.000067 m3 to 0.015 ± 0.012 m3. There was no significant difference in growth between the transplants in the NC and C treatments (p > 0.05) (Fig. 2). From 11 to 18 mo, plants did not grow further and even lost volume to an average of 0.0080 ± 0.0067 m3, although this volume loss was not significant (p > 0.05) (Fig. 2). It should be noted that the plants were irrigated only from 0 to 8 mo and did not receive water for the latter part of the study. Further, during this field trial, the SPRNCA was experiencing a drought. Information from the closest weather station, 8 miles to the northeast in Tombstone, AZ, indicates that the area received a total of 362 mm of precipitation during the 18 mo of this study with approximately 44% of the precipitation occurring in the first 8 mo.


Figure 2
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 2. Average plant volumes for no compost and compost treatments at times 0, 11, and 18 mo where n = 4. Error bars represent one standard deviation.

 
None of the four-wing saltbush transplants survived in the OSC plots, which was unexpected since they are indigenous to the site. Failure to establish transplants in the OSC plots could have been due to a fairly dense growth of indigenous grass. Grasses, like western wheatgrass, have been found to negatively affect four-wing saltbush establishment (Sabey et al., 1990). Thus, it is possible that naturally established grass in this area either out-competed the transplants for nutrients or had an inhibitory effect on their growth.

At the end of the trial, shoot metal contents of transplants were elevated from 2 to 80-fold over levels found in saltbush naturally established in nearby areas (Table 3). The highest levels of metal accumulation were for Mn (~880 mg kg–1), Pb (~310 mg kg–1), Fe (~265 mg kg–1), Al (~230 mg kg–1), and Zn (~200 mg kg–1), reflecting the high total concentration of these metals in the tailings (Tables 2 and 3). There was a positive correlation between the uptake of Fe and the uptake of other metals including Pb, Cu, As, Mn, Al, and Hg (r = 0.89–0.98, Table 4). There were also correlations between uptake of some metals pairs including Pb and As (r = 0.92), Pb and Hg (r = 0.93), As and Hg (r = 0.91), and As and Mn (r = 0.99). However, there was no correlation between soil metal concentrations and plant metal uptake for individual metals (r = –0.22 to 0.45, data not shown).


View this table:
[in this window]
[in a new window]

  		Table 3. Total metal content (mg kg–1) in four-wing saltbush and transplant shoots 18 mo following planting.{dagger}

 

View this table:
[in this window]
[in a new window]

  		Table 4. Pearson's correlation coefficient (r) for metals in four-wing saltbush transplant shoots in tailings treatment plots. Samples were collected at 18 mo.

 
Microbial Analysis
Soil samples were taken from beneath transplant canopies throughout the revegetation trial to monitor changes in the bacterial community. Before the trial began, heterotrophic bacterial numbers in bulk tailings averaged 2.1 ± 0.9 x 105 CFU g–1 dry soil (CFU, colony-forming unit). This is similar to other studies that have reported heterotrophic bacteria in untreated alkaline (pH > 8) tailings (106 CFU g–1) (Shetty et al., 1994; Noyd et al., 1995) and in acidic (pH < 4.5) tailings (105 CFU g–1) (Moynahan et al., 2002). Within 3 mo, a 1 to 1.5 log increase was observed in bacterial numbers under transplant canopies in both the C and NC treatments in comparison to unplanted controls (U). This difference was significant (p < 0.001) and was observed throughout the trial (Fig. 3). It should be noted that the U plots were not irrigated. Despite this difference in irrigation, the enhancement in heterotrophic counts continued when irrigation ceased after 8 mo. Thus, the difference in numbers between planted and unplanted plots was not associated with irrigation effects alone.


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 3. Heterotrophic plate counts beneath transplant canopies for no compost (NC) (circle) and compost (C) (triangle) treatments and for unplanted controls (U) (square) throughout the revegetation trial. Error bars represent one standard deviation (n = 4 for treatments NC and C, n = 2 for the unplanted controls except for *n = 1, **n = 4).

 
Changes in the dominant bacterial populations during the revegetation trial were monitored by DGGE. The PCR products amplified from samples obtained before the trial started were resolved best with a 50 to 60% denaturing gradient. Therefore, all other time points were evaluated using this gradient. The ANOVA analysis showed no significant differences in the total number of bands between the planted treatments and the unplanted controls (p < 0.05) throughout the trial except for 5 mo. At 5 mo, planted community profiles (NC and C) had an average of 30 bands, which was significantly more than for unplanted (U) profiles, which averaged 24 bands (p = 0.02) (Fig. 4). As shown in Fig. 4, the negative control lanes generally had a small number of faint bands. In each case, these bands were subtracted from the profiles during analysis.


Figure 4
View larger version (136K):
[in this window]
[in a new window]

  		Fig. 4. Denaturing gradient gel electrophoresis (DGGE) profiles of PCR-amplified 16S rDNA fragments from tailings treatment plots 5 mo after transplants were established. Labels on top of the gel indicate the sample identity: C, compost plot; NC, no compost plot; U, unplanted control; M, marker ladder composed of heterotrophic mine tailings isolates; and N, negative control.

 
Examination of the banding patterns in each of the DGGE profiles shows that the bacterial community changed with time for each treatment including the unplanted controls. This is not surprising since samples were taken during a 13 mo period that had large moisture and temperature variations (for precipitation a total of 288 mm was measured in nearby Tombstone, AZ, from 31 Oct. 2003 to 31 Nov. 2003, ranging from 0 to 60 mm mo–1; for temperature the high was 38.7°C, the low was –6.8°C, with an average of 20.5°C; these data are from www.wunderground.com; verified 2 May 2007). Dice similarity coefficients were determined for each treatment plot using binary sets of profiles as a function of time. This analysis indicated that the largest changes in community profiles occurred in the first 5 mo of the study for most of the plots (i.e., the biggest difference in % similarities occurred between 0 and 3 mo or between 0 and 5 mo, data not shown).

The DGGE profiles were compared using KNMDS to determine treatment and time effects on the bacterial communities in NC, C, and U plots (Fig. 5). Each KNMDS graph illustrates the first two out of three or four dimensions. In these graphs, the relative distance between samples indicates the amount of similarity or difference among data points.


Figure 5
View larger version (31K):
[in this window]
[in a new window]

  		Fig. 5. Kruskal's Non-Metric Multidimensional Scaling (KNMDS) analysis of denaturing gradient gel electrophoresis (DGGE) profiles from tailings treatment plots as a function of time. Symbols include no compost (NC) (circle), compost (C) (triangle), and unplanted control (U) (square). For each plot, solid lines delineate clusters that are significantly different from each other and dashed lines delineate clusters that show a trend but are not significantly different from each other. The 0 mo plot represents samples taken before transplants were established (stress factor = 0.013, D = 3) (D, dimension) and shows three clusters that are significantly different (p = 0.06). The 3 mo plot (stress factor = 0.026, D = 3) shows no significant clusters. The 5 mo plot (stress factor = 0.069, D = 3) shows two clusters that are significantly different from each other (p = 0.02). The 7 mo plot (stress factor = 0.071, D = 3) shows two clusters that are not significantly different (p = 0.18). The 11 mo plot (stress factor = 0.026, D = 3) shows three significantly different clusters (p = 0.01) and the 13 mo plot (stress factor = 0.023, D = 4) shows three clusters (p = 0.41).

 
The KNMDS analysis of DGGE profiles suggests that there were shifts in the dominant bacterial populations associated with the different treatments (NC, C, U) during the trial. The bacterial communities that developed under plant canopies (NC or C) became more similar to each other and more different from unplanted controls with time (Fig. 5). Before planting, the bacterial community in bulk soil in the tailings treatment plots was divided into three groups or clusters that were quite random (p = 0.06) (Fig. 5, Time–0 Months). This plot indicates that the bacterial communities in treatment plots that clustered together are more similar to each other than to the communities that were not in the cluster. Three months after initiation of the revegetation trial, the original relationships (clusters) had disappeared and no significant clusters were observed (Fig. 5). However, by 5 mo, the profiles from samples taken under transplant canopies became more similar to each other, regardless of treatment (NC or C), than to the unplanted controls (p = 0.02). Seven months following planting the same trend was observed, although the differences were not statistically significant (p = 0.18), and thus clusters are shown with dashed trend lines. At 11 mo, three significantly different clusters were found (p = 0.01). Although planted treatments were divided into two different clusters, is not possible to distinguish between treatments since there are samples from compost and no compost treatments in both clusters (Fig. 5, Time–11 Months). In addition, it can be seen that the relative distance between the two clusters containing communities in planted treatments is smaller than the distances between each cluster and the unplanted controls. Therefore, the communities under transplant canopies were still more similar to each other than to the unplanted controls continuing the trend first observed at 5 mo. The final sampling for DGGE analysis took place at 13 mo. Two additional unplanted control samples were collected at this time, U3 and U4. The KNMDS analysis of this time point again shows the trend (not significant, p = 0.41) that the communities in the four unplanted controls are more similar to each other than to any of the transplanted samples.

To determine whether the community structure under transplant canopies was influenced by the compost that the transplants were originally grown in, the compost DGGE profile was compared to a time series (0, 3, 5, and 11 mo) of DGGE profiles from two treatment plots, NC-1 and C-1 (Fig. 6). The KNMDS analysis shows that the compost community is significantly different from all profiles except the NC1 before planting (0 mo) (p = 0.03).


Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 6. Kruskal's Non-Metric Multidimensional Scaling (KNMDS) analysis of the denaturing gradient gel electrophoresis (DGGE) profile from the compost used to germinate and grow the transplants (compost only) and the profiles from one of the no compost plots (NC1) and one of the compost plots (C1) at times 0, 3, 5, and 11 mo. The plot (stress factor = 0.021, D = 4) (D, dimension) shows two significantly different clusters (p = 0.025).

 

   Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
Conclusions
 REFERENCES
 
Revegetation
This study demonstrates the feasibility of establishing four-wing saltbush transplants in moderately impacted mine tailings sites with minimal inputs of water (a total of 44 L plant–1) and no fertilizer or organic C amendments. Four-wing saltbush transplants showed a 30-fold increase in plant volume in the first 11 mo of this study and then stopped growing. It is not clear why growth stopped. Water stress may have been a factor since irrigation was terminated after the monsoon rains began in summer 2004 (after 8 mo) and the monsoon delivered only 110 mm of precipitation (measured in Tombstone, AZ) during July, August, and September 2004. Metal toxicity is also a possible factor in the observed growth inhibition (Hagemeyer, 1999), given that there were elevated levels of shoot tissue metal concentrations in the four-wing saltbush transplants, which will be discussed further below. These results suggest that long-term monitoring (e.g., >1.5 yr) is necessary to assess if saltbush transplants can establish an enduring and successful vegetative cover.

In regard to phytotoxicity, Table 2 shows that several metals in the Boston Mill site exceed reported soil plant toxicity levels including Pb (500 mg kg–1), As (15 mg kg–1), Cd (3 mg kg–1), Hg (3 mg kg–1), Cu (200 mg kg–1), Zn (400 mg kg–1), and Mn (3000 mg kg–1) (Kataba-Pendias and Pendias, 2001; Munshower, 1993; Mulvey and Elliott, 2000). However, no toxicity studies have been performed specifically with four-wing saltbush so it is not clear what levels of soil metal concentrations cause toxicity in this plant. While it was beyond the scope of this study to explore the mechanism of metal uptake into shoot tissue, we did make the following observations. Metal uptake was not attributed to total soil metal concentrations because there was no correlation between soil metal concentration and plant metal uptake (r = –0.22 to 0.45). However, there was a positive correlation between Fe uptake and uptake of other metals including Pb, Cu, As, Mn, Al, and Hg (r = 0.89 to 0.98, Table 4). The exception to this observation was Zn uptake, which was not correlated with the uptake of Fe or any other metal. It is possible that root exudates from plants were responsible for solubilizing the metals from the soil. At the alkaline pH of the tailings used in this study, Fe is not soluble and therefore is unavailable for plant uptake. Plants adapted to alkaline soils have been shown to solubilize Fe by releasing H+ from their roots (Brown, 1976). This effectively lowers the pH in the root zone, which enhances solubilization of Fe as well as other metals.

Of all the metals present, only Pb was translocated into shoot material in levels that exceed the guidelines for metal toxicity limits for domestic animals, specifically cattle. A recent National Research Council report (National Research Council, 2005) states the limit for Pb is 100 mg kg–1. In this study, measured Pb levels in four-wing saltbush transplant shoot tissues exceed this guideline by up to threefold (Table 3). A comparison of available data indicates that Pb uptake may be saltbush species dependent. For example, while four-wing saltbush has been reported to take up Pb from acidic mine tailings in greenhouse experiments (Sabey et al., 1990), other studies indicate that two other Atriplex species, A. numularia, and A. lentiformis, did not shoot accumulate Pb from alkaline and acidic tailings, respectively (Williams et al., 1994; Jordan et al., 2002; Mendez et al., 2007). The observed shoot accumulation of Pb in this study is undesirable as wildlife may be negatively impacted. Thus, other native species that naturally establish in the tailings area, such as big sacaton grass or other Atriplex species, should be investigated.

Microbial Community Analysis
One way in which the introduction of plants into mine tailings sites improves ecosystem function is by enhancing microbial activity, which in turn enhances soil productivity and supports plant establishment and succession. Thus, we hypothesized that following planting the presence of root exudates and compost C inputs would lead to an increase in culturable heterotrophic bacterial numbers in planted treatments. This hypothesis was confirmed with a 1 to 1.5 log increase in heterotrophic numbers (CFUs) in planted treatment plots. This increase occurred whether or not compost was added, suggesting that the plants, rather than added compost, enhanced the heterotrophic bacterial community beneath transplant. canopies. This is further supported by the data showing that the compost community DGGE profile was quite different from the NC and C community profiles (Fig. 6). Note that both of these communities had exposure to the compost. For the NC treatments the plants were grown in compost before transplanting and for the C treatments the plants were both grown in the compost and were amended with compost during transplanting.

We also hypothesized that plant establishment would lead to an increase in species richness in the Boston Mill site. This hypothesis was not confirmed. An ANOVA analysis comparing the number of bands among DGGE profiles suggests that there was no significant increase (p > 0.05) in richness between planted areas and the unplanted controls (except at 5 mo, p = 0.02). While DGGE profiles do not provide total richness indices, as one band may represent more than one species (Jackson et al., 2000), these profiles can serve as a screening tool to observe gross differences in species richness between multiple samples. Interestingly, some studies have suggested that root exudates can serve as selective compounds decreasing richness indices rather than increasing them (Marilley et al., 1998; Kozdrój and van Elsas, 2000).

As just discussed, an increase in heterotrophic bacterial numbers is not equivalent to an increase in species richness, nor is it equivalent to an increase in community functional diversity (e.g., nutrient cycling, response to environmental stress). In fact it has been suggested that heterotrophic bacterial numbers recover more readily than functional diversity during reclamation (Moynahan et al., 2002). This may be due to a slower response in the bacterial genetic pool. Although species richness is not equivalent to functional diversity, there is some indication that there may be a critical level of species richness below which function is impaired (Griffiths et al., 1997). This is an important issue because low functional diversity might compromise the ability of the bacterial community to normally degrade organic matter, and thus, affect nutrient cycling. One impact would be an accumulation of organic matter in the site. This question can only be resolved by long-term monitoring of the site to investigate if there is organic matter accumulation.

Spatial heterogeneity and seasonal changes often makes it difficult to find significant differences among treatments in field studies. In this study, although KNMDS results were not significant for every time point, a consistent trend was observed regarding bacterial community structure from 5 mo on. In general, DGGE profiles from planted treatments were more similar to each other than to the unplanted controls. In contrast, there was no consistent difference in profiles taken beneath transplant canopies in compost and no compost treatments (similar to the plant growth results). Thus, KNMDS analysis suggests that the establishment of plants, in either the absence or presence of compost amendment, caused a similar change in the bacterial community structure beneath transplant canopies. Several other groups studying bacterial community structure have also observed community changes during plant growth (e.g., Marschner et al., 2002; Kang and Mills, 2004; Lerner et al., 2006). It would be intriguing to determine whether the community structure changes observed in this study could be used to develop biomarkers to predict the stage and success of phytostabilization in arid and semiarid mine tailings sites.

Temporal analysis of unplanted control DGGE profiles indicates that shifts occurred in the bulk soil bacterial community (e.g., the Dice similarity coefficient for time 0 and time 13 mo was 62%). The magnitude of this shift is actually quite similar to that found for the NC (58.5%) and the C (52.6%) treatments suggesting that there are natural seasonal changes in the bacterial community structure in both the control and treated plots. Such temporal shifts in soil communities have also been observed previously in biological soil crusts (Nagy et al., 2005) and in an upland grassland soil in the UK (Griffiths et al., 2003). In the latter study, profiles obtained from 5 to 10 cm and from 10 to 15 cm changed more than profiles obtained from 0 to 5 cm. In the present study, samples were obtained from approximately 5 cm.

The results of this study along with similar findings from other groups show that natural temporal variations in the bacterial community can affect the interpretation of results. For example, there was a single time point, 5 mo, where we observed significant differences in the number of bands among the DGGE profiles of unplanted and planted areas. If this was the only time point analyzed, this finding would have suggested that there were increases in bacterial diversity associated with planted areas but not unplanted controls. Therefore, to assess reclamation success through microbial community changes it is important to monitor changes at various times throughout the year.


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
REFERENCES
 
Previous work has shown that reclamation strategies such as liming, seeding, and compost addition can increase heterotrophic numbers and change community structure in both acidic and alkaline mine tailings (Noyd et al., 1995; Mummey et al., 2002; Moynahan et al., 2002; Kelly et al., 2003). Here we have shown that the establishment of four-wing saltbush transplants in the Boston Mill mine tailings site caused similar changes in the bacterial community as detected by heterotrophic counts and DGGE profile analysis. In this moderately impacted mine tailings site, compost amendment was not essential to either establishment of plants or to the development of the bacterial community. Study results also suggest that increased heterotrophic bacterial abundance and shifts in the dominant bacterial populations did not translate into an increased bacterial species richness. Long-term monitoring needs to be performed to investigate whether the observed changes are sustained over time and to determine whether the observed shifts in community structure reflect potential increases in functional diversity. In summary, the practical implication of this work is that reclamation efforts in alkaline mine tailings should take into consideration that the addition of compost may not provide any benefit for plant establishment. This should be a cost-reducing advantage for implementing phytostabilization in moderately impacted mining sites.


   ACKNOWLEDGMENTS
 
This research was supported by Grant 2 P42 ES04940-11 from the National Institute of Environmental Health Sciences Superfund Basic Research Program, NIH. Our thanks to Karl Ford and William L. Auby of the Bureau of Land Management for their support and cooperation in giving us access to the Boston Mill Site. We also wish to thank Michael Kopplin of the University of Arizona Superfund Basic Research Program Hazard Identification Core for performing all ICP-MS total metal analyses.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
G. Qu, A. de Varennes, and C. Cunha-Queda
Use of Insoluble Polyacrylate Polymers to Aid Phytostabilization of Mine Soils: Effects on Plant Growth and Soil Characteristics
J. Environ. Qual., December 30, 2009; 39(1): 168 - 175.
[Abstract] [Full Text] [PDF]

Home page
 Appl. Environ. Microbiol.Home page
S. L. Iverson and R. M. Maier
Effects of Compost on Colonization of Roots of Plants Grown in Metalliferous Mine Tailings, as Examined by Fluorescence In Situ Hybridization
Appl. Envir. Microbiol., February 1, 2009; 75(3): 842 - 847.
[Abstract] [Full Text] [PDF]

Home page
 Am J Trop Med HygHome page
T. J. J. Inglis, A. Levy, A. J. Merritt, M. Hodge, R. McDonald, and D. E. Woods
Melioidosis Risk in a Tropical Industrial Environment
Am J Trop Med Hyg, January 1, 2009; 80(1): 78 - 84.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Agricola
Right arrow Articles by Rosario, K.
Right arrow Articles by Maier, R. M.
Related Collections
Right arrow Ecosystem Restoration
Right arrow Bioremediation and Biodegradation
Right arrow Heavy Metals

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome