Gene Expression Changes in Arabidopsis thaliana Seedling Roots

Arabidopsis thaliana root transcriptome responses to the munition, hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), were assessed using serial analysis ...
0 downloads 0 Views 169KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 6313-6320

Gene Expression Changes in Arabidopsis thaliana Seedling Roots Exposed to the Munition Hexahydro-1,3,5-trinitro-1,3,5-triazine DREW R. EKMAN,† N. LEE WOLFE,† AND J E F F R E Y F . D . D E A N * ,‡ National Exposure Research Laboratory, Ecosystems Research Division, U.S. Environmental Protection Agency, Athens, Georgia 30605, and Daniel B. Warnell School of Forest Resources, University of Georgia, Athens, Georgia 30602

Arabidopsis thaliana root transcriptome responses to the munition, hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), were assessed using serial analysis of gene expression (SAGE). Sequencing of SAGE libraries from control and RDXexposed root tissues revealed induction of genes known to respond to a variety of general stresses. Among the highly induced genes were several encoding molecular chaperones and transcription factors as well as vacuolar proteins and peroxidases. Strongly repressed transcripts included ones encoding ribosomal proteins, a cyclophilin, a katanin, and a peroxidase. Comparison of the transcriptional profile for the RDX response to a profile previously described for Arabidopsis roots exposed to trinitrotoluene (TNT) revealed significant differences in the inferred gene expression patterns. This suggests that Arabidopsis employs drastically different mechanisms for coping with these two compounds. With respect to the goal of engineering plants to better tolerate and degrade explosives at contaminated sites, these results suggest that enhancement of different genes and metabolic pathways may be required to deal effectively with each type of explosive. This has ramifications for phytoremediation efforts since many contaminated sites harbor both compounds.

Introduction The polynitramine explosive, hexahydro-1,3,5-trinitro-1,3,5triazine, also known as Royal Demolition eXplosive or RDX (Figure 1), is an environmental contaminant of concern at munitions manufacturing and disposal facilities across the U.S. and Europe (1). Long recognized for its neurological effects in humans (2-4), acute toxicity levels of RDX and its impacts on fecundity have been established for a variety of mammals (ATSDR 1995) as well as birds (5), fish (6), and invertebrates (7, 8). RDX has also been classified by the U.S. EPA as a class C (possible) human carcinogen on the basis of limited feeding studies in mice. Although it has limited solubility in water, RDX is relatively mobile in soils and groundwater (9, 10) and can accumulate in plants grown in contaminated soils (11). On one hand, the uptake of RDX by plants is of concern for the potential entry of this contaminant into the food chain * Corresponding author phone: (706)542-1710; fax: (706)542-8356; e-mail: [email protected]. † U.S. Environmental Protection Agency. ‡ University of Georgia. 10.1021/es050385r CCC: $30.25 Published on Web 07/14/2005

 2005 American Chemical Society

FIGURE 1. Chemical structures for munition contaminants (TNT and RDX) discussed in this study. These nitrogenated munitions are highly recalcitrant and toxic compounds that are found at high levels in the soil as well as in ground and surface water, at or near many military installations. (12, 13), but it also opens the door to cost-effective treatment of contaminated sites through phytoremediation techniques (14). Thus, RDX uptake has been examined in various plant species having characteristics suitable for phytoremediation processes, including bush bean (12), poplar trees (15), and a variety of aquatic and wetland plant species (11). Transformation and mineralization of RDX has been reported in certain plant tissue explants and extracts (11, 16, 17) suggesting that phytoremediation has the potential for complete removal of this contaminant. However, some reports on phytoaccumulation suggest limited transformation of RDX in other plant species (12, 18), and several studies have reported on RDX phytotoxicity (13, 19, 20). Whether for the sake of improving plant tolerance to RDX or for enhancing the inherent transformation and degradation capacities of plants for RDX, there is a need for additional understanding of the metabolic responses plants have to RDX exposure. We previously used serial analysis of gene expression (SAGE) (21, 22) to generate gene expression profiles for Arabidopsis seedling roots cultivated in the presence and absence of trinitrotoluene (TNT) (23). Comparison of these profiles identified genes and pathways induced by TNT and suggested what biochemical mechanisms might be open to manipulation for improving the plant’s ability to detoxify this nitroaromatic compound. In this study, the same technique was used to create a gene expression profile for Arabidopsis seedling roots exposed to RDX, and changes in gene transcript profiles are compared and contrasted to those found in roots exposed to TNT. Despite perceived similarities between these two munitions Arabidopsis gene expression patterns in response to exposure to these two compounds were starkly different. This suggests that efforts to develop plants suitable for phytoremediation of sites contaminated with both compounds may need to address multiple and distinct metabolic pathways.

Materials and Methods Plant Material, Growth Conditions, and Root Tissue Isolation. The following experimental descriptions were developed to conform as closely as possible to the MIAME (Minimum Information About a Microarray Experiment) guidelines being pushed as good practice for all transcriptional profiling experiments (24). A. thaliana ecotype Columbia seeds (WT2, Lehle Seeds, Round Rock, TX) were surface-sterilized by immersion in 70% (v/v) aqueous ethanol for 2 min, followed by immersion in a solution of 30% (v/v) aqueous bleach (ca. 24 mM NaClO final concentration) containing 0.2% Triton X-100 for 40 min with gentle shaking. After washing with several changes of sterile water, the sterilized seeds were placed in sterile Murashige and Skoog (MS) liquid medium prepared according to the manufacturer’s instructions (InVOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6313

vitrogen Life Technologies, Carlsbad, CA). Plants (ca. 40) were grown in 500 mL Erlenmeyer flasks containing 200 mL of this liquid medium for 14 days at 25 °C under a 16-h photoperiod with constant shaking at 85 rpm in a growth chamber. RDX (>98% purity) was obtained from the U.S. Army Center for Environmental Health Research (Fort Detrick, MD). Toxicity of the compound for Arabidopsis was assessed by adding RDX from a stock solution in dimethyl sulfoxide (DMSO) to yield final concentrations of 0, 5, 20, 40, 80, 100, and 150 mg/L in MS medium. Triplicate cultures of Arabidopsis seedlings at each of these RDX concentrations were returned to the growth chamber for 5 days, during which time the seedlings in each flask were examined for signs of stress (leaf chlorosis and necrosis). A final concentration of 150 mg/L RDX was judged to produce notable stress in the plants without causing death, so root tissues for SAGE library construction were isolated from seedlings grown 14 days in liquid MS medium prior to dosing with RDX to this final concentration. Control tissues were isolated from seedlings grown under the same conditions and dosed with an equivalent volume of DMSO. The seedlings were grown in the presence of RDX or DMSO for 24 h, after which time they were submerged briefly in deionized H2O to remove excess medium, and excised roots were immediately frozen in liquid nitrogen. All root tissues were stored at -80 °C prior to RNA extraction. The root tissues for each SAGE library were pooled from approximately 1400 seedlings recovered from 36 treatment flasks. Prior to seeding, the 36 flasks were divided into three replicates of 12 with growth of each replicate being started on a different day. RNA Isolation and cDNA Synthesis. Total RNA from root tissues was extracted using the LiCl precipitation technique of Chang et al. (25) Polyadenylated RNA was isolated from total RNA using Dynabeads oligo-dT(25) magnetic beads (Dynal Biotech, Lake Success, NY) at a ratio of 0.25 mg of total RNA per 250 µL of Dynabead suspension and following the manufacture’s instructions. Double-stranded cDNA was synthesized from 5 µg of poly(A) RNA using the Superscript Choice cDNA synthesis kit (Invitrogen Life Technologies) and following the manufacturer’s protocol, except that a 5′biotinylated oligo dT(18) primer was used in the first-strand reaction. SAGE Library Construction. SAGE libraries were constructed according to the SAGE Detailed Protocol, Version 1.0c (22), a brief description of which follows. Biotinylated cDNAs from each tissue sample were bound to streptavidincoated magnetic beads and digested with NlaIII, a restriction enzyme recognizing the four-base sequence, CATG (anchoring enzyme). DNA released by this digestion was washed away, and the beads, with the adherent 3′ ends of each cDNA, were split into two pools. Linkers containing a binding site for BsmF1 (a Type-II restriction endonucleasesthe tagging enzyme), but different sites for polymerase chain reaction (PCR) primers, were ligated to the NlaIII cleavage site at the 5′ ends of the bead-bound cDNA fragments in each pool. Both pools of cDNAs were digested with BsmF1 to release SAGE tags from the beads, after which the pools were combined, and 102 bp linker-flanked ditags were formed by blunt-end ligation. Following amplification of the ditags by PCR using Hybaid PCR Express thermocyclers (Thermo Electron Corp., Milford, MA) equipped with 96-well blocks, the linkers were removed by NlaIII digestion, and ditags were ligated to form concatemers. The concatemers were subsequently size-fractionated (700-1200 bp), ligated into the pZero vector (Invitrogen Life Technologies), cloned, and sequenced. DNA sequencing was performed using an ABI 3700 capillary electrophoresis DNA sequencer (Applied Biosystems, Foster City, CA) to analyze products generated using the Big Dye Terminator system (Applied Biosystems) as per the provided protocols. 6314

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005

SAGE Data Analysis. Sequence files were compiled and analyzed using the SAGE Software, v. 3.03, provided by Dr. Kenneth Kinzler (Johns Hopkins University, Baltimore, MD). Tags containing linker sequences and repeated ditags were excluded prior to analysis. Because the library representing control roots was sequenced to a slightly greater extent (31 973 tags for the control versus 31 209 tags for the RDX treatment), values for the RDX library tags were normalized prior to making comparisons of relative gene expression. Ratios were used to compare the relative expression of tags between the two libraries, and in instances where a particular tag was absent from a library, a value of 1 was substituted to avoid division by zero. Using the SAGE software, Monte Carlo simulations were performed to estimate the statistical significance of any differential expression. The null hypothesis for these analyses was that the abundance, type, and distribution of transcripts were the same in both libraries. Gene Identification. To identify the genes from which the tags were derived, each 10-base tag plus the 4-base NlaIII recognition sequence was first compared against the Arabidopsis Gene Initiative (AGI) database of model genes using the Patmatch analysis tool available on The Arabidopsis Information Resource (TAIR) server (http://www.arabidopsis.org). If the tag was found to match exactly the NlaIII site closest to the 3′ end of a model gene transcript, this identity was accepted for the tag. Tags that could not be found in the model gene database were compared against all Arabidopsis sequences in GenBank using the same Patmatch tool. Exact matches were annotated accordingly. Quantitative PCR. Total RNA from Arabidopsis plants grown under conditions identical to those used to generate RNA for the SAGE studies was used to independently verify the expression of selected genes using quantitative RT-PCR. Messenger RNA isolated using Dynal Oligo dT25 magnetic beads served as a template for single-stranded cDNA synthesis using Taqman Reverse Transcription Reagents (Applied Biosystems, Foster City, CA). Real-time fluorescent detection of RT-PCR products was performed using an iCycler iQ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA) and Sybr Green PCR Master Mix (Applied Biosystems). The following PCR primers were designed for this study using Primer Express v. 1.0 (Applied Biosystems): putative peroxidase (At1 g49570) forward, 5′-TCCGTGGATTCGAAGTCATTG-3′ and reverse, 5′-GCAACTATGTCAGCGCATGAA-3′; 60S ribosomal protein L34 (At1g26680) forward, 5′-AAAGTAGCCCCCAAGGCTTAAG-3′ and reverse, 5′-GTCCTCACATGACTGCACCAA-3′; and putative HMG protein (At2g17560) forward, 5′-TGGTAAGGCTGCTGGAGCTAG-3′ and reverse, 5′-GCTCTCAGCCTTAGCGACGT-3′. Amplification reactions were carried out according to the manufacturer’s specifications as follows: 2 min at 50 °C followed by a 10 min activation of the enzyme at 95 °C, and 40 subsequent cycles consisting of 95 °C for 15 s followed by 60 °C for 1 min. All amplification reactions were run using a dilution series of cDNA from both control and RDX-treated tissues in the same PCR master mix. Amplimers were checked for purity and size by gel electrophoresis to ensure that the correct sequence was amplified. All samples were treated with DNAse prior to reverse transcription using DNA-free (Ambion, Inc., Austin, TX) to remove any contaminating genomic DNA. Control reactions omitting reverse transcriptase were run for all samples to ensure that genomic contamination did not contribute to the amplified products. Melt curve analysis confirmed the absence of both primer-dimer and nonspecific product formation.

Results Toxicity of RDX for Arabidopsis thaliana in Liquid Culture. Sterile Arabidopsis seedlings were grown for 14 days in liquid Murashige and Skoog medium before treatment with RDX

TABLE 1. SAGE Library Statistics frequency distributiona 19 to 5

g20

b

4 to 2

)1

totals

unique tags tags sequenced

153 (1.3)b 7063 (22.6)

RDX Library 1021 (8.4) 8498 (27.2)

3039 (25.1) 7739 (24.8)

7909 (65.2) 7909 (25.4)

12 122 31 209

unique tags tags sequenced

178 (1.4) 7276 (22.8)

Control Library 1005 (7.9) 8219 (25.7)

3214 (25.3) 8156 (25.5)

8322 (65.4) 8322 (26.0)

12 719 31 973

a Frequency distributions were calculated based on the total number of unique or sequenced tags in each library shown in the totals column. The percentages of tags in each frequency group are shown parenthetically.

TABLE 2. SAGE Tags Induced at Least 8-Fold by Exposure to RDX tag sequence

tag abundance fold increasea control RDX RDX/control p-chanceb

locus

annotation ribosomal RNA? DnaJ11 heat-shock protein gamma-VPE (vacuolar processing enzyme) ribosomal RNA? peroxidase, putative 3-dehydroquinate synthase, putative ribosomal RNA? H+-transporting ATPase 16K chain P2, vacuolar alpha-hydroxynitrile lyase-like protein intergenic region between At4g29710 and At4g29720 cyclic-nucleotide phosphodiesterase Myb family transcription factor proline-rich membrane protein with G protein-coupled receptor motifs aquaporin (plasma membrane intrinsic protein 1B) SPX-domain protein sulfite reductase heat shock protein expressed protein of unknown function possible protein folding/sorting

GTGGTAACGG TGCTTACCGT GGATAACATC GGTTAGTCGA CGCTGACATA TTCAAGTCCA GCCGTTCTTA TCTTCTCGAA GTGATGCTCT TCCCCTATTA GGAGACAGTG GATGTCTGGC TTTGGAGGAG

1 1 1 1 1 1 3 1 1 1 1 1 2

30 14 14 13 12 11 32 10 10 9 9 9 17

30.0 14.0 14.0 13.0 12.0 11.0 10.7 10.0 10.0 9.0 9.0 9.0 8.5

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.00

2 hits At4g36040 At4g32940 2 hits At1g49570 At5g66120 2 hits At4g38920 At5g10300 1 hit At4 g18930 At1g35516 At4g19200

TTCCTATTCT TTTCTGGATA TAATGTAATG CCTGGAATCA ATCCTTGTCT AGAAAGCAGG

2 1 0 1 1 1

17 8 8 8 8 8

8.5 8.0 g8.0 8.0 8.0 8.0

0.00 0.02 0.00 0.02 0.02 0.02

At2g45960 At5g20150 At5 04590 At5g12030 At2g37110 At5g35200

a For the calculation of expression ratios, a value of 1 was substituted where tag counts were zero. b p-chance values are averages of three Monte Carlo simulations.

at concentrations ranging from 5 to 150 mg/L. Visual signs of toxicity (mainly leaf chlorosis and necrosis) were not evident in any of the cultures, except at the highest dosage (150 mg/L) after 72 h of exposure. While this concentration was in excess of the aqueous solubility of RDX (approximately 40 mg/L) it was nonetheless necessary to induce signs of toxicity within a time frame comparable to that used in the previous study of Arabidopsis gene responses to TNT (23). This suggests that the insoluble fraction (approximately 110 mg/L) over time entered solution as the amount initially in solution was removed by the plants. This would imply that RDX must accumulate to high levels within the plant tissues before signs of toxicity become apparent. SAGE Libraries. SAGE libraries representing the transcripts expressed in Arabidopsis root tissues grown in the presence or absence of RDX were sequenced until just over 30 000 tags were characterized from each condition. Theoretically, each unique tag represents a unique mRNA transcript from a specific gene. Consequently, the terms unique tag and unique gene are essentially synonymous and will be used interchangeably here. The complete SAGE data sets for this study is available from two online sources: as Supporting Information to this paper (http://pubs.acs.org/) or from the Gene Expression Omnibus (GEO) at NCBI (http:// www.ncbi.nlm.nih.gov/geo/) as experimental series GSE560. Table 1 summarizes the frequency distribution of unique tags found in the control and RDX-challenge libraries. Among the 31 209 tags characterized from the RDX-challenge library, 12 122 representing unique genes were identified, of which

7909 were seen only once (singletons). For comparison, 12 719 unique sequences were found among the 31 973 tags characterized in the control library, with 8322 representing singletons. The percentages of unique tags falling into different abundance classes were comparable with values seen in previous studies of plant gene expression using SAGE (23, 26). Differentially Expressed Transcripts. In response to addition of RDX to liquid culture medium, 135 Arabidopsis genes (tags) were induced at least 5-fold in root tissues compared to control roots. Moreover, a relatively large number of tags (6672) were observed only in the RDXchallenge library. Among the induced tags were many representing transcripts associated with general plant stress responses, such as heat-shock proteins and chaperonins as well as other functional classes, including defense-related proteins, transporters, transcription factors, and vacuolar proteins (Table 2). The most highly induced tag (GTGGTAACGG) as well as two others among the seven tags most highly induced by RDX exposure (GGTTAGTCGA and GCCGTTCTTA) serves to highlight one weakness SAGE has as a transcriptional profiling technique. None of these tags falls within a currently annotated gene, even though each exists in numerous Arabidopsis ESTs found in GenBank (111, 36, and 29 ESTs for each of the three tags, respectively). Each of the three actually corresponds to a different conserved domain within the ribosomal RNA complex, which is duplicated on chromosomes 2 and 3 of A. thaliana. Thus, as a consequence of gene VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6315

FIGURE 2. An unannotated Arabidopsis transcriptional unit identified using SAGE. A browser view of results from the Arabidopsis tiling array transcriptome express tool (27) for the genome region corresponding to the 10th tag on the list of tags most induced in Arabidopsis roots by exposure to RDX (TCCCCTATTA). The position and orientation of the tag sequence on the Arabidopsis genome is noted at the bottom of the figure by the arrowhead labeled my query. The histogram in the center of the figure indicates transcriptional signals detected by the tiling array for overlapping 25-nucleotide sequence windows corresponding to either the top or bottom strand of genomic DNA. duplication, none of these three SAGE tags can be uniquely ascribed to a single sequence in the genome. A similar situation was also noted for a highly abundant tag (CGCCCGCCGC) detected in our study of gene responses in Arabidopsis roots exposed to TNT (23). That tag, which was induced 5-fold by TNT exposure, but only 3-fold by RDX exposure, matched yet another conserved domain in the duplicated rRNA domains of chromosomes 2 and 3. While duplicated genes present obvious problems for interpretation of SAGE data, one strength SAGE has, in contrast to closed-system transcriptional profiling techniques, such as DNA microarrays, is the ability to identify novel transcriptional units. As was the case for the tags corresponding to rRNA domain sequences, the tag that was tenth most-induced by RDX exposure (TCCCCTATTA) also did not correspond to an annotated Arabidopsis gene. However, in contrast to the rRNA-derived sequences, the sequence for this tag only occurs once in the A. thaliana genome, at a position on chromosome 4 that does not correspond to any previously recognized transcriptional units or ESTs. As shown in Figure 2 though, data from A. thaliana tiling arrays (27) indicate that at least one very short transcript is produced in some abundance from this part of the genome. It, therefore, seems likely that the gene corresponding to this tag has not previously been recognized because its product is short and typically discarded from cDNA pools by the sizefractionation steps that are a routine part of cDNA library preparation. Moving to tags that could be mapped to specific and recognized Arabidopsis genes, the tag that was the second most highly induced by RDX-treatment corresponded to a DnaJ-like protein. Proteins in this family are known to act 6316

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005

on their own as molecular chaperones or as co-chaperones in association with Hsp70 (28-30). Induction of this gene suggests that the A. thaliana root cells exposed to RDX were experiencing stress-related problems with protein folding and/or stability. A vacuolar-processing enzyme (γ-VPE), a type of protease, was also induced by RDX exposure. Proteases of this type serve a variety of cellular functions (31), including assisting in the maturation of defense proteins (32, 33) and activation of enzymes involved in senescence and programmed cell death (34, 35). The strong induction (12x) of a putative peroxidase transcript may indicate oxidative stress in the plant roots or, alternatively, a mechanism for oxidation of RDX. The fungus, Phanerochaete chrysosporium, is known to degrade the nitroaromatic explosive, TNT, as well as a number of other organic pollutants using oxidative enzymes, such as peroxidases. Degradation of RDX by this fungus has been reported to proceed by an oxidative mechanism, presumably catalyzed by lignin peroxidases (36). Relatively little is known of the metabolic function of 3-dehydroquinate synthase, and regulation of its expression in plants is even less well understood, but it was induced in response to pathogen-derived elicitors (37). Vacuolar ATPase subunit genes respond to a variety of stress conditions, and the one identified as up-regulated in this study (VHA-c3) is typically expressed in root caps and responds to drought or salt stress (38). Among the 10 tags most induced by RDX exposure, the one corresponding to an R-hydroxynitrile lyase-like protein (GTGATGCTCT) may bear special notice. These enzymes, which in crucifers such as Arabidopsis are often most highly expressed in germinating seeds and in defense responses, catalyze the cleavage of hydroxynitrile compounds. Presuming a mechanism for hydroxylation the nitro moieties of RDX,

TABLE 3. SAGE Tags Repressed at Least 8-Fold by Exposure to RDX tag sequence

tag abundance fold increasea control RDX control/RDX p-chanceb

locus

annotation

GTTCTGCAAA TTTAAAAAAA TCTGAAAGAG TTTCACACTT TTGGTGTTTC GCCCTGCGAT AAAGGCGGCG GGACAGATTC GGAAATGAAC GATCGACCAA ATCATCATCA

13 11 11 10 10 10 10 9 9 9 9

1 1 1 1 1 1 1 0 0 1 1

13.0 11.0 11.0 10.0 10.0 10.0 10.0 g9.0 g9.0 9.0 9.0

0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.00 0.00 0.01 0.01

multiple hits multiple hits multiple hits At4g30170c At5g63030 BU917433 At2g17560 At5g62300 multiple hits At1g20310 multiple hits

TAATTATGTA TAAGAAATCT TAAAAGCTTT GTCTTAATGA GGAAATAAAA GACGCTAGCG ATGTGTTGGT ATCAAAAAAA AGGAAGGAAG

8 8 8 8 8 8 8 8 8

0 1 1 0 1 1 0 0 1

g8.0 8.0 8.0 g8.0 8.0 8.0 g8.0 g8.0 8.0

0.00 0.02 0.02 0.00 0.02 0.02 0.00 0.00 0.02

At4g39675 At4g17530 At2g47320 At1g02900 At1g27030 At2g10410 At2g34560 AK175294 2 hits

chromosomes 2 & 5 (numerous ESTs) chromosomes 2 & 5 (numerous ESTs) chromosomes 1, 3, & 5 (numerous ESTs) peroxidase ATP8a glutaredoxin-like protein small mRNA EST (chromosome 4) putative HMG protein ribosomal protein S20 chromosomes 1, 2, 4, & 5 (numerous ESTs) expressed protein (chloroplastic?) penultimate CATG At4g27840 (expressed protein)? (numerous genome hits in both orientations - -SSRs) expressed protein ras-related small GTP-binding protein RAB1c putative peptidyl-prolyl cis-trans isomerase rapid alkalinization factor (RALF) family protein expressed protein expressed protein putative katanin full-length EST At3g52920 (expressed protein); At5g03660 (expressed protein)

a For the calculation of expression ratios, a value of 1 was substituted where tag counts were zero. b p-chance values are averages of three Monte Carlo simulations. c Likely splice variant.

TABLE 4. Comparison of Transcript Expression Levels as Determined by SAGE and qPCR control

RDX

ratio RDX/control

Amplicon/(model gene)

qPCRa

SAGE

qPCR

SAGE

qPCR

SAGE

putative peroxidase (At1g49570) 60S ribosomal protein L34 (At1g26880) putative HMG protein (At2g17560)

1.1 ( 0.2 100 ( 5.1 57.1 ( 0.1

1 15 10

8.51 ( 0.01 91.9 ( 5.4 10.5 ( 2.4

12 15 1

7.70 0.92 0.18

12.00 1.00 0.13

a

Average values for duplicate qPCR measurements ( the SD.

as appears to occur when plants deal with nitroaromatic compounds, such as TNT (39, 40), induction of an R-hydroxynitrile lyase may hint at the metabolic pathway Arabidopsis attempts to invoke to deal with RDX exposure. More than 100 tags were repressed at least 5-fold in response to RDX exposure, and Table 3 lists all the tags that were at least 8-fold more abundant in control roots than they were in treated roots. Unfortunately, several of the tags on this list corresponded to sequences found at multiple points in the Arabidopsis genome and also mapped to ESTs derived from multiple genes. Among the tags that could be mapped to specific genes, decreased expression was noted for a tag representing a glutaredoxin-like protein. Glutaredoxins typically act as shuttles for reducing equivalents needed to reactivate oxidized proteins and metabolites, such as dehydroascorbate. In Arabidopsis, the glutaredoxin gene family comprises at least 31 members, but none has yet been ascribed a firm physiological function (41). Another tag strongly repressed by RDX corresponded to an HMG (high mobility group) protein. Members of this class of proteins (five have been identified in Arabidopsis) are thought to act as architectural elements in chromatin structure, and through their impact on chromatin are suspected to influence such processes as transcription and recombination (42, 43). Overall, however, there was significantly less interpretable information to be gleaned from the list of genes whose expression was repressed by RDX exposure. Quantitative PCR. SAGE has previously been shown to provide an accurate reflection of gene expression levels for medium- and high-abundance transcripts, but the induction or repression of selected Arabidopsis transcripts in response

to RDX exposure was independently verified in this study using real-time quantitative PCR. Genes that SAGE analysis suggested were induced (At1g49570; putative peroxidase), repressed (At2g17560; HMG protein), or remained relatively unaffected (At2g46880; 60S ribosomal protein L34) by RDX exposure were tested. As shown in Table 4, the quantitative PCR data were in general agreement with the SAGE data. Differences in Gene Expression Between TNT- and RDXExposed Arabidopsis Roots. In a study of gene expression changes induced in A. thaliana roots by 2,4,6-trinitrotoluene (TNT), SAGE identified a variety of highly induced transcripts that corresponded with gene products, such as cytochrome P450s, known for their involvement in TNT metabolism (23). Comparison of TNT- and RDX-treatment results revealed significant differences in the gene responses in Arabidopsis roots to these two compounds. Table 5 lists the expression levels for cytochrome P450 genes induced by one or both of these munitions. Six genes encoding cytochrome P450 enzymes were induced by TNT, but not RDX, including a putative CYP81D11 transcript whose expression was induced at least 14-fold by TNT. RDX exposure but not TNT significantly induced two genes encoding cytochrome P450 enzymes, while both munitions induced a single cytochrome P450 gene. SAGE tags for a great many genes that were strongly induced by TNT exposure were not found in RDX-treated root tissues. Among these were those for several glutathione S-transferase transcripts, including the one most highly induced by TNT exposure (27x) as well as transcripts for such oxidative stress enzymes as monodehydroascorbate reductase and dehydroascorbate reductase (see the SupVOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6317

TABLE 5. Differential Expression of Cytochrome P450 Genes in Arabidopsis Roots in Response to TNT or RDX Exposure

tag sequence

a

tag abundance control TNT

RDX

fold increase for RDXa RDX/TNT

locus

classification

AGAGAAAGTG GCTGAGAGAC GAAAAGATTT AATAAAACTT AAGCATCCGC ACACCAAAGC

0 0 0 2 0 2

14 5 5 9 4 11

TNT-Induced Cytochrome P450s 0 e0.07 0 e0.20 0 e0.20 1 0.11 1 0.25 3 0.27

At3g28740 At4g22690 At2g30750 At2g24180 At1g64950 At1g78490

CYP81D11 CYP706A2 CYP71A12 CYP71B6 CYP89A6 CYP708A3

TATGCCGCCC ATTGCGCGTG

1 1

0 1

RDX-Induced Cytochrome P450s 5 g5.00 5 5.00

At1g16400 At3g20940

CYP79F2 CYP705A30

ACCTAACTGA

0

Munition-Induced Cytochrome P450 5 5 1.00

At4g13310

CYP71A20

For the calculation of expression ratios, a value of 1 was substituted where tag counts were zero.

FIGURE 3. Diagrammatic representation of SAGE tag distribution between treatments. The unique tag distribution in each sage library (control, TNT, or RDX) as well as the overlapping distribution of tags seen in multiple libraries is depicted in this Venn diagram. porting Information). Thus, while metabolism of TNT in plants probably requires a multiphase oxidative mechanism, as was described previously (23), it appears that relatively little overlap exists with the metabolic pathway for RDX. A comparison of SAGE tags identified under all three treatment conditions (control, TNT-, and RDX-exposure) is depicted in Figure 3. Some 3572 unique tags were detected under all three conditions, and most represented housekeeping genes necessary for seedling growth under these culturing conditions. However, many tags were only detected in a single library, with the most (2185) coming from the control tissues.

Discussion Much of what we know about the metabolic pathways plants utilize to cope with xenobiotic agents was inferred from studies using traditional biochemical and molecular biological techniques, which are typically limited to the simultaneous analysis of no more than a handful of relatively abundant enzymes. Genomic technologies make it possible to quickly analyze thousands of gene products in parallel and detect relatively subtle changes in gene expression responses under varying growth conditions. The complete genome sequences that currently exist for three plant species (Arabidopsis, rice, and poplar) also make it possible to identify previously undetected genes that may respond to the unusual metabolic 6318

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005

challenges posed by xenobiotic compounds. Consequently, the use of model plant systems and functional genomic techniques has the potential to have rapid and profound impact on how plants are selected and improved to meet unusual environmental demands, such as those encountered in the development of phytoremediation processes. In this study, a transcriptional profiling technique, SAGE, was used to examine the gene expression responses in Arabidopsis roots exposed to RDX, a munition of some concern for its widespread distribution as an environmental contaminant. SAGE generates short stretches of DNA sequence, or “tags”, that are essentially unique for every gene expressed in a given biological sample and uses highthroughput DNA sequencing to identify and count these tags when they are strung together in long chains (http:// www.sagenet.org/findings/index.html). The technique has been most widely used in molecular oncology (http:// cgap.nci.nih.gov/SAGE) but is become increasingly appreciated for use in other systems, including plants. A strength of SAGE as a functional genomics technique is its ability to analyze gene responses in any organism, whether substantial gene sequence information for that organism already exists. Although this study employed traditional SAGE, which produces sequence tags that may be too short to use in isolating novel genes under some conditions, modified protocols that produce longer tags are available (44, 45). In the case of unstudied plant species found to have unusual potential for handling specific xenobiotic agents, these modified SAGE procedures would speed the identification and isolation of novel genes that could be used for enhancing phytoremediation or bioremediation processes. With respect to plant responses to RDX, this study demonstrated that the mechanisms for RDX metabolism in Arabidopsis are vastly different than from those the plant uses when exposed to TNT. Although these compounds are often lumped together as nitrogenated explosives, this result was anticipated from previous studies that demonstrated little or no transformation of RDX in different plant species (12, 15, 46). This observation will be a major consideration in the development of plants for phytoremediation of RDX since TNT and RDX are often found together at the same contaminated sites. Our SAGE data as well as the results from prior studies (12, 15, 46) suggest that in order for plants to handle both of these explosives simultaneously they will likely require the introduction of suites of genes specific for each compound. Although little or no metabolism of RDX was detected in some plant systems (12, 15, 46), there are some plant systems that have demonstrated that plants do have enzyme systems that could catalyze RDX breakdown (17). Cytochrome P450

enzymes are often utilized for transformation of cellular toxins, either to make them more amenable to conjugation (e.g., attachment of glutathione or six-carbon sugars) or to increase solubility, and the involvement of cytochrome P450 enzymes has been reported previously in the degradation of RDX by Rhodococcus sp. (47). SAGE analysis demonstrated that three Arabidopsis cytochrome P450s were induced at least 5-fold in roots exposed to RDX. Further work will be necessary to determine whether these cytochrome P450 gene responses signal a potential route to the transformation of RDX or simply a general stress response brought about by RDX accumulating to levels that disrupt normal cellular function. Certainly the observed induction of various chaperonins and proteases in the RDX-exposed roots indicates significant disruption of normal cellular function in these tissues. In normal terrestrial plant systems, RDX in not retained with the roots, like TNT, but is distributed throughout the plant, tending to accumulate in leaves and stems, which are the downstream ends of the transpiration process. Consequently, future studies to examine changes in gene expression in these locations after RDX uptake and accumulation will be important to gaining a full understanding of how plants respond to this compound. Coupled with field studies and quantitative analyses of the xenobiotic agents and their metabolites, transcriptional profiling techniques, such as SAGE, promise to greatly enhance our ability to employ plants as important tools for environmental remediation.

Acknowledgments This work was supported by NNEMS Fellowship U-91587201-0 from the U.S. Environmental Protection Agency to D.R.E. Thanks to Dr. W. Walter Lorenz for his assistance with the SAGE technique and Drs. MacArthur Long and Steven McCutcheon for their continued support of the project. This paper has been reviewed in accordance with the U.S. Environmental Protection Agency’s peer and administrative review policies and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

(8)

(9) (10) (11)

(12)

(13) (14) (15) (16) (17)

(18)

(19)

(20)

Supporting Information Available The complete SAGE data set described in this paper in spreadsheet format. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Haas, R.; Schreiber, I.; Vonlow, E.; Stork, G. Conception for the investigation of contaminated munition plants. 2. Investigation of former RDX-plants and filling stations. Fresenius’ J. Anal. Chem. 1990, 338, 41-45. (2) Vonoettingen, W. F.; Donahue, D. D.; Yagoda, H.; Monaco, A. R.; Harris, M. R. Toxicity and potential dangers of cyclotrimethylenetrinitramine (RDX). J. Ind. Hyg. Toxicol. 1949, 31, 2131. (3) Schneider, N. R.; Andersen, M. E. Toxicity and tissue distribution of cyclotrimethylenetrinitramine (RDX). Toxicol. Appl. Pharm. 1975, 33, 198-198. (4) Testud, F.; Glanclaude, J. M.; Descotes, J. Acute hexogen poisoning after occupational exposure. J. Toxicol., Clin. Toxicol. 1996, 34, 109-111. (5) Gogal, R. M.; Johnson, M. S.; Larsen, C. T.; Prater, M. R.; Duncan, R. B.; Ward, D. L.; Lee, R. B.; Salice, C. J.; Jortner, B.; Holladay, S. D. Dietary oral exposure to 1,3,5-trinitro-1,3,5-triazine in the northern bobwhite (Colinus virginianus). Environ. Toxicol. Chem. 2003, 22, 381-387. (6) Burton, D. T.; Turley, S. D.; Peters, G. T. The toxicity of hexahydro1,3,5-trinitro-1,3,5-triazine (RDX) to the freshwater green-alga Selenastrum capricornutum. Water, Air, Soil Pollut. 1994, 76, 449-457. (7) Peters, G. T.; Burton, D. T.; Paulson, R. L.; Turley, S. D. The acute and chronic toxicity of hexahydro-1,3,5-trinitro-1,3,5-

(21) (22)

(23)

(24)

(25) (26)

(27)

triazine (RDX) to 3 freshwater invertebrates. Environ. Toxicol. Chem. 1991, 10, 1073-1081. Robidoux, P. Y.; Hawari, J.; Bardai, G.; Paquet, L.; Ampleman, G.; Thiboutot, S.; Sudahara, G. I. TNT, RDX, and HMX decrease earthworm (Eisenia andrei) life-cycle responses in a spiked natural forest soil. Arch. Environ. Contam. Toxicol. 2002, 43, 379-388. Selim, H. M.; Xue, S. K.; Iskandar, I. K. Transport of 2,4,6trinitrotoluene and hexahydro-1,3,5-trinitro-1,3,5-triazine in soils. Soil Sci. 1995, 160, 328-339. Singh, J.; Comfort, S. D.; Hundal, L. S.; Shea, P. J. Long-term RDX sorption and fate in soil. J. Environ. Qual. 1998, 27, 572577. Best, E. P. H.; Sprecher, S. L.; Larson, S. L.; Fredrickson, H. L.; Bader, D. F. Environmental behavior of explosives in groundwater from the Milan Army Ammunition Plant in aquatic and wetland plant treatments. Uptake and fate of TNT and RDX in plants. Chemosphere 1999, 39, 2057-2072. Harvey, S. D.; Fellows, R. J.; Cataldo, D. A.; Bean, R. M. Fate of the explosive hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) in soil and bioaccumulation in bush bean hydroponic plants. Environ. Toxicol. Chem. 1991, 10, 845-855. Price, R. A.; Pennington, J. C.; Larson, S. L.; Neumann, D.; Hayes, C. A. Uptake of RDX and TNT by agronomic plants. Soil Sediment Contam. 2002, 11, 307-326. Hannink, N. K.; Rosser, S. J.; Bruce, N. C. Phytoremediation of explosives. Crit. Rev. Plant Sci. 2002, 21, 511-538. Thompson, P. L.; Ramer, L. A.; Schnoor, J. L. Hexahydro-1,3,5trinitro-1,3,5-triazine translocation in poplar trees. Environ. Toxicol. Chem. 1999, 18, 279-284. Just, C. L.; Schnoor, J. L. Phytophotolysis of hexahydro-1,3,5trinitro-1,3,5-triazine (RDX) in leaves of reed canary grass. Environ. Sci. Technol. 2004, 38, 290-295. Van Aken, B.; Yoon, J. M.; Just, C. L.; Schnoor, J. L. Metabolism and mineralization of hexahydro-1,3,5-trinitro-1,3,5-triazine inside poplar tissues (Populus deltoides x nigra DN-34). Environ. Sci. Technol. 2004, 38, 4572-4579. Bhadra, R.; Wayment, D. G.; Williams, R. K.; Barman, S. N.; Stone, M. B.; Hughes, J. B.; Shanks, J. V. Studies on plantmediated fate of the explosives RDX and HMX. Chemosphere 2001, 44, 1259-1264. Burton, D. T.; Turley, S. D.; Peters, G. T. The Acute and chronic toxicity of hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) to the fathead minnow (Pimephales promelas). Chemosphere 1994, 29, 567-579. Winfield, L. E.; Rodgers, J. H.; D’Surney, S. J. The responses of selected terrestrial plants to short (