Characteristic of the enrichment community that degrades 1,3-D
Gas chromatographic analysis showed that after 8 weeks of enrichment, volatilization fluxes (μg h–1) were much lower in the compost-amended soils compared with the unamended soils for the (Z) and (E) isomers of 1, 3-D (Fig. 1). Fluxes were monitored for 96 h each week for the first 8 weeks after injection of fumigant. The average cumulative volatilization losses in terms of percentage of applied 1,3-D showed that the amount of 1,3-D recovered from the amended soil was much lower than that from the unamended soil (Fig. 2a,b). This was in agreement with a previous report by Gan et al. (1998a) that showed lower volatilization fluxes for compost-amended soil compared with unamended soil. Low volatilization means high degradation rates in these samples. Analysis of week by week cumulative volatilization losses as a percentage of 1,3-D applied provided a full understanding of the contribution of organic amendment to the increased rate of degradation. After 8 weeks, degradation of 1,3-D was significantly enhanced compared with the first 4 weeks. At the end of the experiment, about 8% of applied (Z) and 5% of (E) 1,3-D was lost from the amended soils, compared with 20% and 15%, respectively, for the unamended soils.
Figure 1. Volatilization flux (μg h–1) of 1,3-dichloropropene isomers from columns packed with the Arlington sandy loam, with and without organic amendment, after Telone-11 injection at 30 cm depth. Point differences indicate 4 h time intervals between sampling, and the data points indicated in the graph are for an 8-week sampling period. Cis=Z isomer and trans=E isomer. (●), cis; (○), trans; (▮), C-cis; (□), C-trans
Download figure to PowerPoint
Figure 2. Cumulative volatilization losses in percentage of applied 1,3-D. (a) Z isomers from amended and unamended soils, and (b) E isomer from amended and unamended soils. (●), Amended soil, cis 1,3-D; (○), unamended soil, cis 1,3-D
Download figure to PowerPoint
The degradation rate of both (Z) and (E) isomers of 1,3-D in amended soil after 6 months showed first-order kinetics, with the constant K 0·14 or half-life (t½) of 2 days (Fig. 3a,b). This suggests that 1,3-D will be volatilized into the air by gas-phase diffusion within 2 days at this rate of degradation. Degradation of 1,3-D in the unamended soil was substantially reduced, with a first-order t½ of 5·7 days for the (Z) isomer and 4·9 days for the (E) isomer. When both the amended and the unamended soils were autoclaved, the half-life for the (Z) isomer increased to 6 days for the amended soil and 14 days for the unamended soil. For the (E) isomer, the half-life increased to 5·3 days for the amended and 8·75 days for the unamended soil. It took an additional 3 days to degrade both isomers of 1,3-D without organic amendment, suggesting the contribution of organic amendment to the degradation of 1,3-D. Seven more days were required for the degradation of 1,3-D in the sterilized unamended soil (E) isomer, and 12 days for the (Z) isomer, suggesting that a longer time is required for chemical degradation to remove 1,3-D from the environment.
Figure 3. Degradation rates of 1,3-D isomers in compost-amended and unamended Arlington sandy loam soil and enrichment culture. (a) Z isomer (b) E isomer. (●), Amended soil; (○), unamended soil; (▾), mixed colonies; (▮), unamended sterilized soil; (▪), sterilized amended soil
Download figure to PowerPoint
Since the focus of this study was to investigate bacterial populations that may be representative of long-term fumigant-adapted microbial communities involved in the degradation process, the cultivable populations that can grow on basal mineral medium with 1,3-D as the sole carbon source were initially observed. The enrichment culture with 1,3-D as the sole carbon source was used to study degradation of this compound. Also, sterilized soils, and soils with and without amendment, were used to compare degradation with the enrichment culture. The enrichment culture was unable to degrade the (Z) (t½ of 3·7 days) and (E) (t½ 3·4 days) isomers of 1,3-D as fast as the amended soil after 4 days of incubation (Fig. 3a,b). It is interesting to note that the enrichment culture was more effective in transforming the (E) isomer (80%) than the (Z) isomer (65%).
Total degradation of 1,3-D was higher in the amended soil than the unamended soil and significantly lower in the sterile soils. The 4 day incubation study confirmed the observations in Figs 1 and 2 that the degradation capacity of these samples was predominantly of biological origin. The enrichment culture was plated on solid basal mineral media with 1,3-D as the sole carbon source, and separate colonies were tested to determine their capacity to degrade 1,3-D. None of the tested colonies, either individually or in a combination of two or three, was capable of degrading 1,3-D. This shows a plating bias towards selecting 1,3-D-degrading populations. This is why there is no report of isolation of 1,3-D-degrading populations from soil that has not been repeatedly treated with the compound for many years. In fact, most of the reports of 1,3-D-degrading isolates are from soils that have been treated with the compound for about 5–10 years. Verhagen et al. 1995 demonstrated that repeated treatment of soil with 1,3-D for about 10 years resulted in accelerated microbial degradation of this compound. They isolated 15 bacterial strains with 1,3-D-degrading capacity. Our results clearly demonstrate that compost-amended soils can significantly enhance degradation of 1,3-D from soil with no history of 1,3-D application. However, this process is carried out by a consortium of micro-organisms that is very difficult to isolate. One possible reason for this may be that during the first year of 1,3-D treatment of agricultural soils, most of the biological degradation is a result of direct metabolism, co-metabolism or fortuitous metabolism.
Comparison of 1,3-D-degrading microbial communities by 16S rDNA DGGE
DGGE patterns of amended and unamended soil, with and without 1,3-D (Fig. 4a), were analysed to determine differences in banding patterns which may reflect differences in microbial communities adapted to the fumigant and also associated with degradation of this fumigant. Comparison of the bacterial communities showed the banding patterns to be very similar between 1,3-D-amended soils (lanes 4 and 5). However, the banding pattern from the control soil with 1,3-D (lane 3) was different from the amended soil. Differences in banding patterns were also observed in the amended soil without fumigant (lane 2) when compared with those with fumigants. This showed that some species in the fumigant-treated soil became more prevalent. Seven major bands were detected in the 1,3-D-treated soils (E1, E2, E3, E4, E5, E6 and E8), and these bands represented detectable strains of the microbial communities that had been formed during the six-month period of the experiment. The significance of this may be that the bacterial strains from these bands may degrade 1,3-D. Bands E1, E2, E3, E5, E6, E7, E8, E9 and E14 were dominant bands in the two 1,3-D-treated soils (Fig. 4a, and see Table 1 for names of different clones and their accession numbers). Bands E9 and E12 were unique to the umamended soil with fumigant, and these clones were mainly unidentified species.
Figure 4. DGGE analysis of bacterial communities from compost-amended and control soils. Amplified products were separated on a gradient of 30–70% denaturant. All labelled bands were excised from the gel, re-amplified and subjected to sequence analysis. These re-amplification products were cloned, and the clones were screened as described in the text. (a) Community structures after six months of enrichment. Lane 1: Arlington sandy loam without organic amendment; lane 2: with organic amendment; lane 3: without organic amendment but 1,3-D; lanes 4 and 5: with organic amendment and 1,3-D. (b) Community structures for duplicate jars of 1,3-D after serial dilution with phosphate buffer from 10–1 to 10–7. The numbers 1–7 indicate 10–1–10–7. (c) Community structures after serial dilution. No new bands were developed from these treatments. The banding patterns of organic-amended (CM) and control samples were different from each other, with the CM treatment with the control soil having more diverse banding patterns
Download figure to PowerPoint
Table 1. Micro-organisms identified from predominant 16S rDNA DGGE bands obtained from compost-amended and unamended 1,3-dichloropropene-fumigated soil
The technique of amplifying the 16S rDNA gene from the soil community avoids the bias imposed by selection but introduces other unknown factors. Eichner et al. (1999) showed that the number and intensity of bands do not equal the number and abundance of species within the microbial community, due to features of 16S rDNA-based phylogeny and bias inherent to PCR amplification from complex template mixtures. The reasons for most of the limitations are that DGGE banding patterns are subjected to PCR bias due to DNA extraction methods, potential preferential amplification and the formation of chimera (Wintzingerode et al. 1997). Other problems may result from one organism producing more than one DGGE band because of multiple, heterogeneous rRNA operons (Cilia et al. 1996; Nübel et al. 1996; Rainey et al. 1996). Also, for some phylogenetic groups of bacteria, 16S rDNA sequences do not allow discrimination between species, so one DGGE band may represent several species with identical rDNA sequences (Vallaey et al. 1997).
In the second analysis, 1,3-D-fumigated soils amended with compost manure and non-fumigated samples were serially-diluted in minimal media to 10–7. DGGE analysis was done on these samples after DNA extraction. Compost manure only and control soils were also diluted to 10–4; the soils were enriched with 1,3-D as the sole carbon and the same analysis was carried out. This was done to uncover species that may not be revealed by the banding pattern due to the complexity of the samples. The main bands of 16S rDNA amplified from compost-amended 1,3-D-enriched soil were still visible after diluting 10–3. Two new bands appeared at 10–3 and were excised as clone E15 and E16 (Fig. 4b). Band E16 re-appeared at higher dilutions (10–6), which indicates that these sequences were less efficiently amplified in the highly complex soil samples. No new bands were recovered from compost manure and the control soils at higher dilutions (Fig. 4c). Greater numbers of dominant bands were present in the compost sample than in the control soil. The uncovering of new banding patterns is typical of what happens in microbial communities when equally-abundant species are diluted and the new communities are recovered in less complex communities. Serial dilution was used here to reveal less-abundant species for a proper understanding of the complexity of the enriched communities.
Banding patterns in the amended soil without 1,3-D (Fig. 4c), and after serial dilution as revealed by PCR-DGGE of 16S rDNA of the major microbial communities, showed that this soil was different from the non-amended and compost soils. This clearly shows that the introduction of an organic amendment into agricultural soils adds not only nutrients and other components to the soil that may enhance the structure of the soil, but also new microbial populations. In areas where fumigants are applied, the addition of an organic amendment with the application of fumigants may increase the total quality of that soil (Hoitink and Fahy 1986). Hoitink and Fahy showed that when organic wastes are composted, successional changes in microbial populations and community structure occurs. In the present experiment (Fig. 4a,c), soil without organic amendment (control) (lane 1) had 14 dominant bands, soil with organic amendment (lane 2) had 21 dominant bands, control soil with 1,3-D (lane 3) had 17 bands, and 1,3-D-amended soils (lane 4 and 5) had 19 and 18 bands, respectively. The presence of bands E2, E3, E5, E6, E7, E8 and E9 in 1,3-D-amended soil, but not in the control soil, suggested that more bacterial strains became adapted to 1,3-D. The results in this study suggest that some micro-organisms in soils treated with compost manure can adapt to 1,3-D and form microbial consortia that can degrade this compound.
Analysis of DGGE bands
To investigate the dominant organisms in the compost-amended fumigant-treated soils, fragments from 16 selected DGGE bands were sequenced and subjected to BLAST Gene Bank analysis as shown in Table 1. According to the BLAST description, all our clones were members of the gamma subdivision of Proteobacteria and Gram-positive bacteria, with a few clones classified as unidentified soil bacteria. Bands E3, E4 and E15 were Gram-positive bacteria found in 1,3-D-amended soil. The partial sequence from clone E4 found in this group showed it to be present in the four soils with 1,3-D and organic amendment. This suggests the dominance of Actinomadura species in compost-amended samples. This species was first isolated from tropical rain forest soils by Wang et al. (1999). Bands E8, E1, E14 and E16 were dominated by Cytophaga and Pseudomonas species. Clone E1 had a 100% similarity index with Ps. reactans and this clone was found in all the fumigant-treated soils. Pseudomonas sp. may be one of the most abundant bacterial species in soil, and there are many reports of the ability of this species degrade 1,3-D (Lebbink et al. 1989; Verhagen et al. 1995). Fifteen bacterial strains with the capacity to degrade 1,3-D (of which four were Pseudomonas sp.) were isolated from enrichment cultures grown from adapted soils (Verhagen et al. 1995). One of the strains, Ps.cichorii 170, was shown to degrade 1,3-D completely to 3-chloroallyl alcohol (Poelarends et al. 1998). These authors noted that genetic characterization of the genes involved in the complete metabolism of 1,3-D was identical to the dhaA of the Gram-positive bacterium Rhodococcus rhodochrous NCIMH13064 and the dhiA genes from Rhodococcus sp. strain m15–3 (Bosma et al. 1999). In the present study, clone E15 was identified as R. ruber, with 99% similarity to the 16S rDNA sequence of this bacterium. This clone was identified after serial dilution of our samples. Mycobacterium sp. (clones E3 and E5), Saccharomonospora viridis (clone E2) and Nocardiopsis sp. (clone E6) were identified as major bands in treatments with 1,3-D, suggesting that these bacteria may be involved in the degradation of 1,3-D.
The importance of this information is that once a community is established, it is possible that it will continue to use 1,3-D as the sole carbon source for growth and other metabolic activities, resulting in the accelerated degradation of this compound. The major advantage of this approach is the ability to identify and classify fumigant-adapted or -degrading microbial populations without the requirement for isolation of pure culture. As it is well established that the isolation of degrading populations from soil can be a very long process, DGGE can be used as a quick method for screening soils for the major isolates. With this information, different culturing media can be used to select the strains that may be involved in degradation. From the present experiments, most of the strains in Table 1 that became dominant after 6 months of microcosm enrichment were of the closely-related Pseudomonas and Actinomadura. In most cases, classical enrichment tends to yield Pseudomonas strains and related Gram-negative bacteria because of their high growth rates. In this study, Gram-positive bacteria became the predominant bacterial populations after prolonged enrichment in the microcosm.