APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 2009, p. 2969–2972 0099-2240/09/$08.00⫹0 doi:10.1128/AEM.02051-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 75, No. 9
TaqMan Probe-Based Real-Time PCR Assay for Detection and Discrimination of Class I, II, and III tfdA Genes in Soils Treated with Phenoxy Acid Herbicides䌤† Jacob Bælum1 and Carsten S. Jacobsen1,2* Geological Survey of Denmark and Greenland (GEUS), Department of Geochemistry, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark,1 and Faculty of Life Sciences, Department of Natural Sciences, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark2 Received 4 September 2008/Accepted 19 February 2009
Separate quantification of three classes of tfdA genes was performed using TaqMan quantitative real-time PCR for 13 different soils subsequent to mineralization of three phenoxy acids. Class III tfdA genes were found to be involved in mineralization more often than class I and II tfdA genes. the relative composition of the different classes, the need for a tool to do this is evident. Therefore, the main objective of the present study was to develop a reliable quantitative real-time PCR-based method to separately quantify the different classes of tfdA genes directly in complex soil samples. Two novel qPCR assays (tfdA 81-bp and TaqMan assays) and one previously described qPCR assay (tfdA 215-bp assay) (1) targeting the known diversity of tfdA genes were tested to determine their PCR efficiency and specificity for the different classes of tfdA genes. Due to potential problems with production of PCR artifacts, the 215-bp PCR fragment is in the longer range of the fragment size recommended for qPCR (18), explaining why we designed a novel primer set amplifying a shorter fragment. The tfdA 81-bp and tfdA 215-bp assays were based on the SYBR green dye, while the TaqMan assay was based on TaqMan probes. Also, in the TaqMan assay our novel 81-bp primer set was used to amplify the target sequences. Primers for the novel tfdA 81-bp assay (Table 1; see Fig. S1 in the supplemental material), as well as probes for the TaqMan assay, were designed based on tfdA gene sequences obtained from the GenBank database (accession numbers M16730, U25717, and AF377325 for tfdA classes I to III, respectively). To confirm that no unwanted targets are amplified with our primer sets, the oligonucleotides were used as queries for a BLAST search in the GenBank database (for details see the supplemental Materials and Methods in the supplemental material). In the present study we had access to instrument facilities that detect only two fluorophores simultaneously. Therefore, we were not able to test multiplex PCR, where all three tfdA gene classes can be quantified in a single PCR vessel, but we believe that this should be possible by using probes with three different fluorophores. Using 10-fold standard dilution series of recombined plasmids with inserts of each of the class I, II, and III tfdA genes and 107 to 100 genes per reaction as a template, highly comparable PCR performances were achieved with the three different assays (see Fig. S2 and Table S2 in the supplemental material). Except for the tfdA 215-bp assay targeting the class II gene, consistent and reliable quantification was obtained down to a limit of 102 genes per reaction for all three tfdA classes (see Fig. S2 in the supplemental material). The com-
The phenoxy acids (PA), including 2,4-dichlorophenoxyacetic acid (2,4-D), 4-chloro-2-methylphenoxyacetic acid (MCPA), and 2-(4-chloro-2-methylphenoxy)propanoic acid (MCPP), are herbicides that are intensively used for control of broadleaf weeds in cereal crops worldwide. Extensive research on these environmentally hazardous compounds has produced detailed information on pathways and gene sequences involved in their complete mineralization in several bacterial pure-culture isolates (3, 12, 15, 16). Decomposition of the ether bond resulting in a phenolic compound and acetic or propanoic acid is accepted to be the first step in the degradation pathway (5). This step is catalyzed by an ␣-ketoglutarate-dependent dioxygenase encoded by the tfdA or tfdA-like genes (6). The diversity of the tfdA genes has been investigated in detail, and three different classes have been proposed based on sequence information (10). Recently, the tfdA genes have been used as biomarkers in studies of the growth of degraders during mineralization of PA in natural soil samples in situ (1, 2, 7, 8, 19), but none of these studies successfully differentiated quantities of the individual tfdA gene classes. Previously, two different PCR assays were used to detect and quantify tfdA genes in environmental samples. Vallaeys et al. (17) developed a PCR assay suitable for targeting and proving the presence of the three classes of tfdA genes, while, in order to improve specificity and PCR efficiency, Bælum et al. (1) developed and used a novel PCR primer set more suitable for quantitative real-time PCR (qPCR) (1, 2, 11). Furthermore, based on endpoint analysis of PCR products, these two PCR assays have been used to study the functional diversity and dynamics of the different classes of tfdA genes during mineralization of phenoxyacetic acids in environmental samples. As this kind of analysis does not provide a quantitative measure of
* Corresponding author. Mailing address: Geological Survey of Denmark and Greenland (GEUS), Department of Geochemistry, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. Phone: 45 3814 2313. Fax: 45 3814 2050. E-mail:
[email protected]. † Supplemental material for this article may be found at http://aem .asm.org/. 䌤 Published ahead of print on 27 February 2009. 2969
2970
BÆLUM AND JACOBSEN
APPL. ENVIRON. MICROBIOL. TABLE 1. Primers and probes used for qPCR
Oligonucleotide
Type
tfdA-215bp
Forward primer Reverse primer Forward primer Reverse primer Probe Probe Probe
tfdA-81bp tfdA-CI tfdA-CII tfdA-CIII a
Target genes
Sequence (5⬘–3⬘)a
tfdA classes I, II, and III GAGCACTACGCRCTGAAYTCCCG GTCGCGTGCTCGAGAAG tfdA classes I, II, and III GAGCACTACGCRCTGAAYTCCCG SACCGGMGGCATSGCATT tfdA class I FAM-TTGCGCTTCCGAATAGTCGGTGTC-BBQ tfdA class II FAM-CGTTGACTTTCAGAATACTCT GTGTCGCCA-BBQ tfdA class III YAK-TTGACTTTCAGAATAGTCCGTATCGCCAAG-BBQ
Fragment Annealing size (bp) temp (°C)
215
64
81
62 62 62 62
R ⫽ A or G; Y ⫽ T or C; S ⫽ G or C; M ⫽ A or C. FAM, 6-carboxyfluorescein; BBQ, blackberry quencher.
parable qPCR sensitivities for the SYBR green and TaqMan probe qPCR assays are in accordance the sensitivities reported elsewhere (4, 9, 13). Detection (i.e., the replicates were not consistent and thus unreliable quantification was obtained) could be performed with as few as 101 genes per reaction. Probably due to a 1-bp mismatch in the reverse primer region, reliable quantification of class II genes using the tfdA 215-bp assay could be obtained with only as few as 104 genes per reaction (see Fig. S2a in the supplemental material). Furthermore, a test with different combinations of the tfdA target sequences and with the three TaqMan probes together in one reaction vessel was performed in order to verify the specificity of each of the probes for its target sequence. Also, we tested the feasibility of using the probes to perform duplex real-time PCR with combinations of the class I- and III-specific probes and of the class II- and III-specific probes. These tests gave the same results for the detection level and specificity as the tests described above, indicating that duplex PCR is indeed possible with these probes (for detailed information, see the supplemental Materials and Methods in the supplemental material). A soil microcosm experiment including 13 different soils obtained from distinct locations around the world (for further details on soils, see Table S1 in the supplemental material) was performed to test the different qPCR approaches with DNA extracted directly from a wide variety of soils. In order to allow specific degraders harboring the tfdA genes to proliferate, we treated the microcosms with 90 mol of 2,4-D, MCPA, or MCPP kg⫺1 of soil. Prior to and subsequent to ⱖ50% mineralization (measured using evolved 14CO2), a composite subsample consisting of 0.5 g of soil was removed, and DNA was extracted using a Power-Clean soil DNA kit (MoBio Laboratories, Carlsbad, CA) according to the manufacturer’s instructions. Total DNA was quantified and the extraction efficiency was normalized by running 4-l aliquots of extracts on a standard 1.5% agarose gel stained with ethidium bromide, and qPCR was performed using the three different approaches (for further information on the experimental setup, nucleic acid preparation, and qPCR, see the supplemental Materials and Methods in the supplemental material). Despite the low detection limit of the qPCR approaches, we were able to detect low levels of tfdA genes (⬍105 genes g⫺1 soil) in only three soils, SjOreg, Suma-Paz, and KBSreg (for information on soils, see the supplemental Materials and Methods in the supplemental material) prior to PA application, while after mineralization of one of the PA (⬎50% mineralized) we were able to detect significant increases in the levels of tfdA genes in the soils. The potential to mineralize the
three PA was investigated using the widely used and wellvalidated assay for trapping 14CO2 in an NaOH trap during mineralization in microcosms (14). The potential to mineralize ⬎50% of the added PA was found for all 13 soils for 2,4-D, for 9 soils for MCPA, and for 4 soils for MCPP, indicating that the potential for 2,4-D mineralization is more widespread than the potential for MCPA and MCPP mineralization. Additionally, our data suggest that 2,4-D is mineralized more rapidly than MCPA and MCPP (data not shown). In all of the soils exhibiting ⬎50% mineralization we detected increased levels of tfdA genes (Fig. 1), while in the soils with very slow and insignificant mineralization no such increase was detected. Even though there was a slight tendency for the tfdA 81-bp SYBR green and TaqMan assays to reveal larger quantities (Fig. 2A and 2C), the correlation between the tfdA quantities obtained using the three different qPCR approaches was very high (Fig. 2). Based on two-way analysis of variance statistics (P ⬍ 0.05), the only soil scenarios for which a difference between the quantification results could be detected were 2,4-D and MCPA in the Pradera soil and 2,4-D in the Suma-Paz soil and the KBSforest soil. For the Pradera soil treated with 2,4-D the difference was especially noticeable (the outliers are shown in Fig. 2A and 2B). For this soil scenario we were able to find 6 ⫻ 108 tfdA genes g⫺1 soil using the tfdA 81-bp SYBR green assay, while with the two other assays we were able to find only ⬎5 ⫻ 104 tfdA genes g⫺1 soil. This suggests that a degrader harboring a novel class of tfdA genes was present, but due to difficulties in sequencing extremely short PCR amplicons we have not been able to obtain sequence information. We consider the generally high correlation between tfdA quantities obtained using the three qPCR assays strong evidence that these three assays are suitable for consistent and reliable quantification of the class I to III tfdA genes in environmental samples. The usability of our novel TaqMan probe-based qPCR approach was demonstrated by quantifying the three different tfdA gene classes after mineralization of 2,4-D, MCPA, and MCPP in 13 different soils (Fig. 1; see Table S3 in the supplemental material). For most of the soils, we observed that if class III tfdA genes were present in a soil, they typically proliferated and became dominant among the tfdA genes after mineralization occurred. In the soils treated with 2,4-D this was reflected by generally higher ratios of class III tfdA genes to class I tfdA genes (Fig. 1A and 1G). This trend was even more obvious for MCPA and MCPP mineralization, where class I tfdA genes proliferated only in the KBSreg and KBSforest soils treated with MCPA (Fig. 1B), while in the remaining soils
VOL. 75, 2009
CLASS I, II, AND III tfdA GENES IN HERBICIDE-TREATED SOILS
2971
FIG. 1. Ratios of the three tfdA gene classes quantified using the TaqMan assay, expressed as the cumulative numbers of the three classes of tfdA genes in each of the 13 soils after mineralization of 2,4-D, MCPA, and MCPP. (A to C) Class I tfdA genes quantified in soils exposed to the three PA. (D to F) Class II tfdA genes quantified in soils exposed to the three PA. (G to I) Class III tfdA genes quantified in soils exposed to the three PA. The individual values for triplicate samples are indicated, and for clarity, the soils for which data are shown in each panel are indicated at the bottom. The exact numbers are shown in Table S3 in the supplemental material.
showing potential for MCPA and MCPP mineralization only the class III tfdA genes proliferated (Fig. 1B and 1I). Further proof was the fact that the KBSreg and KBSforest soils were the only soils treated with 2,4-D in which class III tfdA genes did not proliferate, suggesting that there was no potential for proliferation of class III tfdA genes, which left the PA for the organisms harboring the class I tfdA genes. In previous studies we speculated that MCPA mineralization is linked to the class III tfdA gene (1) and 2,4-D mineralization is linked to both class I and class III tfdA genes (1, 2). In these studies only two different soils, originating from the same region, were studied. Furthermore, based on enrichment studies of microbial communities, Zakaria et al. (19) linked mineralization of MCPP to the class III tfdA gene. The present study
adds significantly to our knowledge of the dynamics of tfdA genes during mineralization of the PA 2,4-D, MCPA, and MCPP. Here we describe a method to quantitatively measure the relative levels of the different tfdA gene classes in a wide variety of soils, which provides significantly better data. In conclusion, we successfully developed a TaqMan-based qPCR method to quantify three different classes of tfdA genes in environmental samples. Applying the method, we were able to quantify tfdA genes in 13 different soils subsequent to mineralization of 2,4-D, MCPA, and MCPP. In situ in natural soils, mineralization of 2,4-D can induce growth of organisms harboring one of the three tfdA genes, preferentially the class I and III tfdA genes, and the class III tfdA gene is most often the dominant gene. Mineralization of MCPA induces proliferation
2972
BÆLUM AND JACOBSEN
APPL. ENVIRON. MICROBIOL. The Danish Research Council for Nature and Universe is thanked for financial support (FNU grant 274-05-0199). Ziv Arbelli, Shai Arnon, Gary Bending, and Fabrice Martin-Laurent are thanked for kindly providing soil samples. Pia Bach Jakobsen and Szymon Kopalski are thanked for skillful technical assistance, and Kirsa Demant is thanked for proofreading the manuscript. REFERENCES
FIG. 2. Correlations between the numbers of class I to III tfdA genes determined using the three different approaches. (A) tfdA 81-bp assay compared with the tfdA 215-bp assay. (B) TaqMan assay compared with the tfdA 81-bp assay. (C) TaqMan assay compared with the tfdA 215-bp assay. The lines represent linear regressions, and the R2 value is indicated in each panel. The error bars indicate standard errors of triplicate determinations.
of class III tfdA genes, and in cases where the potential for class III gene proliferation is absent, class I genes may proliferate as well. Mineralization of MCPP induces proliferation of only class III tfdA genes.
1. Bælum, J., T. Henriksen, H. C. B. Hansen, and C. S. Jacobsen. 2006. Degradation of 2-methyl-4-chlorophenoxyacetic acid in top- and subsoil is quantitatively linked to the class III tfdA gene. Appl. Environ. Microbiol. 72:1476–1486. 2. Bælum, J., M. H. Nicolaisen, W. E. Holben, B. W. Strobel, J. Sorensen, and C. S. Jacobsen. 2008. Direct analysis of tfdA gene expression by indigenous bacteria in phenoxy acid amended agricultural soil. ISME J. 2:677–687. 3. Bollag, J. M., C. S. Helling, and M. Alexander. 1967. Metabolism of 4-chloro-2-methylphenoxyacetic acid by soil bacteria. Appl. Microbiol. 15: 1393–1398. 4. Demontis, M. A., S. O. Cacciola, M. Orru, V. Balmas, V. Chessa, B. E. Maserti, L. Mascia, F. Raudino, G. M. D. S. Quirico, and Q. Migheli. 2008. Development of real-time PCR systems based on SYBR Green I and TaqMan technologies for specific quantitative detection of Phoma tracheiphila in infected Citrus. Eur. J. Plant Pathol. 120:339–351. 5. Don, R. H., A. J. Weightman, H. J. Knackmuss, and K. N. Timmis. 1985. Transposon mutagenesis and cloning analysis of the pathways for degradation of 2,4-dichlorophenoxyacetic acid and 3-chlorobenzoate in Alcaligenes eutrophus JMP134(pJP4). J. Bacteriol. 161:85–90. 6. Fukumori, F., and R. P. Hausinger. 1993. Alcaligenes eutrophus JMP134 2,4-dichlorophenoxyacetate monooxygenase is an ␣-ketoglutarate-dependent dioxygenase. J. Bacteriol. 175:2083–2086. 7. Gonod, L. V., F. Martin-Laurent, and C. Chenu. 2006. 2,4-D impact on bacterial communities, and the activity and genetic potential of 2,4-D degrading communities in soil. FEMS Microbiol. Ecol. 58:529–537. 8. Lee, T. H., S. Kurata, C. H. Nakatsu, and Y. Kamagata. 2005. Molecular analysis of bacterial community based on 16S rDNA and functional genes in activated sludge enriched with 2,4-dichlorophenoxyacetic acid (2,4-D) under different cultural conditions. Microb. Ecol. 49:151–162. 9. Liu, Z. X., Y. Teng, H. Liu, Y. L. Jiang, X. Y. Xie, H. F. Li, J. Q. Lv, L. Y. Gao, J. Q. He, X. J. Shi, F. Y. Tian, J. S. Yang, and C. X. Xie. 2008. Simultaneous detection of three fish rhabdoviruses using multiplex real-time quantitative RT-PCR assay. J. Virol. Methods 149:103–109. 10. McGowan, C., R. Fulthorpe, A. Wright, and J. M. Tiedje. 1998. Evidence for interspecies gene transfer in the evolution of 2,4-dichlorophenoxyacetic acid degraders. Appl. Environ. Microbiol. 64:4089–4092. 11. Nicolaisen, M. H., J. Baelum, C. S. Jacobsen, and J. Sorensen. 2008. Transcription dynamics of the functional tfdA gene during MCPA herbicide degradation by Cupriavidus necator AEO106 (pRO101) in agricultural soil. Environ. Microbiol. 10:571–579. 12. Pieper, D. H., W. Reineke, K. H. Engesser, and H. J. Knackmuss. 1988. Metabolism of 2,4-dichlorophenoxyacetic acid, 4-chloro-2-methylphenoxyacetic acid and 2-methylphenoxyacetic acid by Alcaligenes eutrophus JMP134. Arch. Microbiol. 150:95–102. 13. Schonenbrucher, H., A. Abdurnawjood, K. Failing, and M. Bulte. 2008. New triplex real-time PCR assay for detection of Mycobactetium avium subsp. paratuberculosis in bovine feces. Appl. Environ. Microbiol. 74: 2751–2758. 14. Sorensen, S. R., A. Schultz, O. S. Jacobsen, and J. Aamand. 2006. Sorption, desorption and mineralisation of the herbicides glyphosate and MCPA in samples from two Danish soil and subsurface profiles. Environ. Pollut. 141: 184–194. 15. Streber, W. R., K. N. Timmis, and M. H. Zenk. 1987. Analysis, cloning, and high-level expression of 2,4-dichlorophenoxyacetate monooxygenase gene tfdA of Alcaligenes eutrophus JMP134. J. Bacteriol. 169:2950–2955. 16. Suwa, Y., A. D. Wright, F. Fukimori, K. A. Nummy, R. P. Hausinger, W. E. Holben, and L. J. Forney. 1996. Characterization of a chromosomally encoded 2,4-dichlorophenoxyacetic acid ␣-ketoglutarate dioxygenase from Burkholderia sp. strain RASC. Appl. Environ. Microbiol. 62:2464–2469. 17. Vallaeys, T., R. R. Fulthorpe, A. M. Wright, and G. Soulas. 1996. The metabolic pathway of 2,4-dichlorophenoxyacetic acid degradation involves different families of tfdA and tfdB genes according to PCR-RFLP analysis. FEMS Microbiol. Ecol. 20:163–172. 18. Wang, X., and B. Seed. 2006. High-throughput primer and probe design, p. 93–106. In M. T. Dorak (ed.), Real-time PCR. Taylor & Francis, New York, NY. 19. Zakaria, D., H. Lappin-Scott, S. Burton, and C. Whitby. 2007. Bacterial diversity in soil enrichment cultures amended with 2-(2-methyl-4-chlorophenoxy)propionic acid (mecoprop). Environ. Microbiol. 9:2575–2587.
