Secondary compound hypothesis revisited: Selected plant secondary metabolites promote bacterial degradation of cis-1,2-dichloroethylene (cDCE)

Cis-1,2-dichloroethylene (cDCE), which is a common hazardous compound, often accumulates during incomplete reductive dechlorination of higher chlorinated ethenes (CEs) at contaminated sites. Simple monoaromatics, such as toluene and phenol, have been proven to induce biotransformation of cDCE in microbial communities incapable of cDCE degradation in the absence of other carbon sources. The goal of this microcosm-based laboratory study was to discover non-toxic natural monoaromatic secondary plant metabolites (SPMEs) that could enhance cDCE degradation in a similar manner to toluene and phenol. Eight SPMEs were selected on the basis of their monoaromatic molecular structure and widespread occurrence in nature. The suitability of the SPMEs chosen to support bacterial growth and to promote cDCE degradation was evaluated in aerobic microbial cultures enriched from cDCE-contaminated soil in the presence of each SPME tested and cDCE. Significant cDCE depletions were achieved in cultures enriched on acetophenone, phenethyl alcohol, p-hydroxybenzoic acid and trans-cinnamic acid. 16S rRNA gene sequence analysis of each microbial community revealed ubiquitous enrichment of bacteria affiliated with the genera Cupriavidus, Rhodococcus, Burkholderia, Acinetobacter and Pseudomonas. Our results provide further confirmation of the previously stated secondary compound hypothesis that plant metabolites released into the rhizosphere can trigger biodegradation of environmental pollutants, including cDCE.

JS666 [13][14][15] . Little additional evidence of direct aerobic cDCE degradation has been reported for isolates affiliated with Acinetobacter sp., Achromobacter sp., Bacillus sp., Pseudomonas sp. 16 or with mixed microbial cultures retrieved from a stream bed sediment 17 and from CE-contaminated groundwater 18 . On the other hand, aerobic microbial degradation of cDCE can occur via aerobic cometabolism (AC), a process driven by oxygenase enzymes which have relaxed substrate specificity 19,20 . In this case, cDCE is fortuitously oxidized in the presence of a growth-supporting substrate that stimulates enzymatic activity of the oxygenase 21 . Growth-supporting substrates, which have been described in relation to cDCE AC processes, are ethene 22 , methane 23,24 , propane 25 , butane 26 , propene 27 , ammonia 28 and some less chlorinated hydrocarbons such as VC 29,30 . In addition, monoaromatic compounds, such as phenol, toluene and benzene, can efficiently enhance aerobic cometabolic mineralization of cDCE 31,32 . Some difficulties with respect to AC implementation arise from the introduction of some of the above-mentioned substrates into the environment mostly due to their toxicity 33 .
Secondary plant metabolites (SPMEs), which enter the soil via different routes, including root exudation and litter decomposition, constitute an important source of nutrients for soil bacteria and play naturally established roles in plant-microbe communication and chemotactic regulation of soil microbial colonization 34,35 . Extensive research has demonstrated that plant-derived compounds also constitute a precious source of promising alternative substrates in order to trigger biodegradative pathways for xenobiotics, including polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs) [36][37][38][39] . This phenomenon is referred to as the "secondary compound hypothesis" 40 . In the present study, we sought monoaromatic SPMEs which could be used to assist the degradation of cDCE, possibly through aerobic cometabolism, in a manner similar to that described in relation to phenol and toluene. To our knowledge, this proof-of-concept study is the first to explore the possibility of cDCE degradation induced by monoaromatic SPMEs, which were used to support microbial growth through a laboratory-enrichment cultivation procedure.

Results
Degradation of cDCE by toluene-and phenol-exposed microbial community. At the initial stage of this study, the microbial community in the contaminated soil was shown not to be capable of cDCE degradation when cDCE was provided as the sole carbon source during the first 10 days of cultivation (Fig. 1). Measurement of cDCE concentrations in the culture headspaces performed up to the third month of cultivation revealed the persistence of cDCE in the liquid cultures (with approximately an 11% of variability among all measurements).
On the other hand, when exposed to toluene and phenol, microbial suspensions extracted from soil successfully degraded cDCE over a period of 10 days (Fig. 1). Higher cDCE depletion levels were achieved in the presence of toluene, with a 82.78% ± 3.73 percentage points (pp) removal of the initial amount despite an initial lag phase of approximately three days (Fig. 1); in the presence of phenol, depletion was 57.89% ± 3.43 pp. Much lower cDCE loss (about 11.8% pp) was observed over the same period of time in phenol-supplied control cultures inhibited with sodium azide (Fig. 1), suggesting that cDCE depletion in toluene-and phenol-supplied cultures was attributable to microbial activity.
Degradation of cDCE by SPME-enriched microbial cultures. The biodegradation of cDCE was first evaluated for eight microbial cultures where the primary carbon sources were represented by different monoaromatic SPMEs. After approximately a month of cultivation, cDCE depletion over a period of 10 days was monitored for all cultures. The highest cDCE depletion levels were attained in the presence of acetophenone (10.14%), phenethyl alcohol (8%), p-hydroxybenzoic acid (9.34%) and trans-cinnamic acid (7.11%). These four cultures were subjected to further enrichment through serial sub-culturing passages, during which substrate depletion, microbial growth and cDCE degradation were evaluated. At the final enrichment stage, all of the SPMEs tested were totally consumed within the first 18 to 25 hours of incubation, despite a lag phase of approximately 10 hours. The measurements of microbial concentrations at the end of the final sub-culturing stage indicated that microbial growth was more effectively supported by trans-cinnamic acid and phenethyl alcohol (Fig. 2). At the same time, significant cDCE depletion was achieved in the presence of acetophenone (88.59% ± 9.08 pp of the initial amount of cDCE), phenethyl alcohol (79.18% ± 13.53 pp), p-hydroxybenzoic acid (69.05% ± 6.30 pp) and trans-cinnamic acid (46.82% ± 4.10 pp) (Fig. 2). When the concentration of cDCE in the liquid phase was plotted over time (Fig. 3), significant cDCE degradation was found to occur after the first day of cultivation, once  the SPME substrates were already consumed, and to slow down considerably after seven days. Negligible cDCE loss (below 5% pp) was observed in sodium azide-inhibited cultures prepared with p-hydroxybenzoic acid as carbon source. Transformation yields of cDCE per SPME, expressed as µmol (cDCE) •mmol (SPME) −1 , were: 16.38 ± 1.55 for acetophenone, 16.32 ± 1.91 for p-hydroxybenzoic acid, 10.53 ± 1.31 for phenethyl alcohol and 9.80 ± 0.76 for trans-cinnamic acid.
Structural diversity of SPME-enriched microbial populations. The highest levels of richness, as assessed by the number of sequence variants observed, or phylogenetic diversity, as assessed by the Shannon, Simpson and reciprocal Simpson diversity indexes, were clearly observed for the initial soil microbial community (Fig. 4). Remarkable differences were observed between the SPME-enriched microbial populations, where the highest diversity indexes were found in acetophenone-enriched cultures, followed by p-hydroxybenzoic acid and phenethyl alcohol-amended cultures, while trans-cinnamic-enriched populations were characterized by the lowest diversity indexes (Fig. 4). In order to highlight the structural differences between the SPME-enriched microbial populations, a non-metric multidimensional scaling (NMDS) analysis was carried out. This showed that the cultures enriched on p-hydroxybenzoic acid and trans-cinnamic acid formed distinct clusters, while cultures enriched on acetophenone and phenethyl alcohol were more related (Fig. 5).
Phylogenetic analysis of the most abundant taxa. The relative abundance profiles of the thirty most abundant sequence variants in the treatments were evaluated. A ubiquitous enrichment of taxa within Proteobacteria and Actinobacteria occurred in the enrichment cultures ( Table 1). The most commonly detected sequence variants amongst Betaproteobacteria clustered with Burkholderiaceae, with Cupriavidus being the most abundant taxon in cultures enriched on trans-cinnamic acid and phenethyl alcohol and quite abundant in cultures enriched on the other SMPEs. Members of the families Comamonadaceae (Simplicispira, unclassified Comamonadaceae) and Oxalobacteraceae (Massilia) were also profusely detected. Gammaproteobacteria were mainly represented by Xanthomonadaceae (genera Dokdonella, Pseudoxanthomonas, Rhodanobacter) and the genera Pseudomonas, Acinetobacter and Legionella. Alphaproteobacteria were more abundant in acetophenone-enriched cultures, which particularly included Devosia and Sphingopyxis. Acetophenone-enriched populations were characterized by a number of additional phylogenetic classes, such as Verrucomicrobiae, Alphaproteobacteria, Flavobacteria, Sphingobacteria, Actinobacteria (Rhodococcus) and Planctomycetia, as well as an enrichment of unclassified bacteria. A significant enrichment of Deltaproteobacteria (mainly Bdellovibrio) was observed in the presence of p-hydroxybenzoic acid and also in the other enriched microbial populations (Table 1). Interestingly, none of the enriched taxa were detected with frequencies of more than 0.001% in the original soil.

Discussion
This proof-of-concept study demonstrates the possibility of cDCE aerobic degradation promoted by the presence of SPMEs used as carbon and energy sources during the enrichment of specific soil microbial populations. The initial microbial community extracted from CE-contaminated soil was shown to be incapable of cDCE biodegradation when exposed to cDCE without the addition of any other carbon and energy source.
Successful cDCE degradation was first achieved upon exposure of the soil microbial community to toluene or phenol ( Fig. 1). A number of studies have demonstrated that cDCE biodegradation is attainable by using aerobic cometabolism with aromatic hydrocarbons such as toluene and phenol 41,42 .
Promising bioremediation potential has often been ascribed to microorganisms associated with plant roots [43][44][45][46] . Some studies have even shown that, in polluted environments, the composition of root exudates can drive a selection of rhizosphere microbial populations with a greater capacity to degrade pollutants 47,48 . Given the structural analogy between toluene, phenol and some monoaromatic SPMEs, we hypothesized that these SPMEs might promote the degradation of cDCE in a manner similar to that for toluene and phenol. Microbial growth on some plant metabolites has already been demonstrated with regard to the cDCE-assimilating bacterium Polaromonas sp. JS666 which is capable of growing on catechol, gentisate and salicylate 14 . Moreover, other plant compounds, including certain types of terpenes, have been used as growth-supporting substrates to enhance microbial degradation of other CEs 49 . The choice of the initially screened SPMEs depended on their monoaromatic structure and on the fact that they are widespread natural compounds produced by different plant species (details of the natural distribution of the SPMEs mentioned in this study are available from the U.S. Department of Agriculture at https://phytochem.nal.usda.gov). This study provides evidence that selected monoaromatic SPMEs can support microbial growth and cDCE depletion. In order to obtain enrichment cultures, the SPME concentrations used in this study were based on previous studies, in which same-class SPMEs were used during research on the cometabolic degradation of other xenobiotic compounds, including PCBs and trichloroethylene 49,50 . At the final stage of enrichment, the presence and type of substrate were found to greatly influence cDCE degradation activity. According to the monitoring data (Fig. 3), cDCE depletion was limited during the first day of cultivation in the presence of acetophenone and zero in the presence of p-hydroxybenzoic acid. This might be due to substrate inhibition phenomena, which are frequently observed during cometabolism 51 . By contrast, cDCE depletion was detectable from the first day of incubation for cultures enriched on phenethyl alcohol and trans-cinnamic acid. In all enrichment cultures, cDCE degradation activity was notable until day 7 of cultivation and decreased considerably thereafter. The overall transformation of cDCE was more effectively driven by acetophenone-, p-hydroxybenzoic acid-and phenethyl alcohol-enriched microbial populations (Fig. 2) compared to that enriched on trans-cinnamic acid. In particular, the highest yields of cDCE removal per SPME used were observed for acetophenone-and p-hydroxybenzoic acid-enriched cultures. It is noteworthy that increased cDCE biodegradation activity took place in acetophenone-and p-hydroxybenzoic acid-enriched cultures, in which lower final microbial concentrations (Fig. 2) were observed along with higher alpha diversity indices for the enriched microbial populations (Fig. 4). In addition, a clear match was found between the increase in alpha diversity and biodegradative activity, expressed as cDCE transformation yields, which is in line with the finding that increased diversity is generally an indication of greater community resilience and is associated with a higher degree of bioremediation efficiency 52 . As highlighted by NMDS analysis, the enrichment of diverse microbial populations was strictly dependent on the type of SPME used. The two-dimensional NMDS plot in Fig. 5 shows how enrichment on p-hydroxybenzoic acid and trans-cinnamic acid shapes the microbial community in a very different way to that on acetophenone and phenethyl alcohol.

Continued
16S rRNA gene-based phylogenetic analysis revealed ubiquitous enrichment of several sequence variants affiliated with microbial taxa, which have been reported to act as degraders of many classes of environmental pollutants. These include members of the following genera: Cupriavidus, previously found to be involved in the degradation of phenolic and chloroaromatic compounds as well as toxic aromatic petroleum hydrocarbons 53 ; Rhodococcus, whose members are notorious for their biodegradation capabilities, also involving chlorinated aliphatic hydrocarbons such as trichloroethylene and VC 54,55 ; Sphingopyxis, with S. chilensis being reported to be capable of growing on certain chlorophenolic compounds 56 ; Sediminibacterium, whose member S. salmoneum was detected in a butane-enriched mixed microbial culture capable of aerobic cometabolic degradation of trichloroethylene 57 ; Dokdonella, detected in the aqueous phase of a model wetland supplied with cDCE-and trans-DCE-contaminated water 58 ; and Pseudomonas, whose members are known to contain di-and monooxygenases which can initiate cDCE oxidation upon induction of cometabolic growth on monoaromatic substrates 59, 60 . In this study, different pseudomonads were detected, including those closely related to P. fulva, P. flavescens, P. benzenivorans and P. argentinensis enriched in acetophenone, phenethyl alcohol and p-hydroxybenzoic acid cultures; and those associated with P.corrugata, P. kilonensis and P. jessenii, enriched in trans-cinnamic acid cultures. Furthermore, enrichment of several unclassified taxa of the family Comamonadaceae occurred in all enriched populations. All the most abundant taxa enriched on SPMEs were mostly below the detection limit in the initial microbial community, which provides further evidence that these taxa were involved in cDCE degradation, as the community initially present in the contaminated soil was not capable of cDCE degradation.
Growth on p-hydroxybenzoic acid was the only condition to raise the abundance levels of Bdellovibrio spp. (Table 1). A previous study has reported the enrichment of members of the family Bdellovibrionaceae in microbial communities, which occurred at a contaminated aquifer after toluene-based biostimulation experiments were performed on the in situ cometabolization of trichloroethylene 61 . However, Bdellovibrio spp. are known to parasitize other Gram-negative bacteria 62 , which may also explain their abundance in SPME-enriched populations. Other microbial sequence variants were exclusively observed in some of the SPME-enriched cultures; this was the case for Acinetobacter johnsonii detected in acetophenone cultures as well as Acinetobacter calcoaceticus and Acinetobacter pittii in trans-cinnamic acid cultures (Table 1). To date, a limited number of Acinetobacter spp. have been shown to participate in the degradation of pollutants such as diesel fuel hydrocarbons via alkane hydroxylase enzymes 63,64 . 16S rRNA gene data analysis demonstrated that the type of SPME remarkably influenced the structure and composition of enriched microbial populations, which was a reflection of diverse cDCE degradative activities. However, the enrichment of microbial genera, such as Cupriavidus, Burkholderia, Pseudomonas, and Rhodococcus (Table 1), throughout the SPME-enriched cultures suggests that microorganisms affiliated with those genera might play a role in the degradation of cDCE.
In conclusion, this study provides evidence of enhanced cDCE degradation by microbial communities selected in the presence of SPMEs. Our results suggest that natural attenuation of cDCE in soil might occur as a result of the activity of plant-associated soil microorganisms. The study of SPME-stimulated microbial degradation of CEs could open up new avenues for the exploitation of SPMEs in biotechnology in areas such as the development of purpose-built phytoremediation techniques. In basic research terms, additional data are required on the fate of the original substrates -both SPMEs and cDCE -including the identity of microbial populations which assimilate them as well as the pathways used for their degradation. This could be done with the aid of parallel stable isotope probing experiments as shown elsewhere 65 . Furthermore, the isolation of cDCE-degrading pure microbial cultures could shed light on cDCE degradation mechanisms, including the molecular basis of SPME action, the genes/enzymes involved and their regulation.

Materials and Methods
Chemicals, culture medium and cultivation conditions. Liquid cDCE (97%) and all the SPMEs (purity higher than 97% for caffeic acid and vanillic acid, while higher than 99% for all the other SPMEs) used were obtained from Sigma-Aldrich (USA). Toluene was purchased from Penta (Czech Republic), and phenol (99.5%) was obtained from Fluka (USA). All other chemicals were of analytical grade. Liquid microbial cultures were prepared in an inorganic chloride-free minimal salt medium (MSM) with the following composition per liter of deionized water: Na 2 HPO 4 •12H 2 O 11.0 g, KH 2 PO 4 2.7 g, (NH 4 ) 2 SO 4  Cultivation on SPMEs. Microbial cultures were separately established in the same way as cultures exposed to toluene or phenol but were amended with one of the following eight monoaromatic SPMEs as sole carbon sources: salicylic acid, acetophenone, phenethyl alcohol, p-hydroxybenzoic acid, trans-cinnamic acid, ferulic acid, vanillic acid and caffeic acid. An initial concentration of 50 mg•L −1 of each SPME (equivalent to molar concentration values between 0.3 and 0.4 mM) was supplied to each culture along with a spike of ∼0.25 µmol of cDCE, corresponding to an initial concentration of ~5 µmol•L −1 of cDCE in the aqueous phase. No biological replicates were used in this initial screening experiment. The cultures described above were kept for a period of 33 days and were newly propagated into fresh medium containing SPME and cDCE each time cDCE was completely depleted. Four microbial cultures exhibiting higher cDCE depletion rates over time were selected for further enrichment on SPMEs and cDCE-degradation experiments.

SPME-based liquid enrichment cultures.
Each culture selected was used to inoculate new enrichment cultures in three replicates (25% v/v into fresh MSM), which were propagated every 10 days. A total of seven propagation steps were performed. At the beginning of each culturing step, each vial was provided with 50 mg•L −1 of each SPME and was spiked with ∼0.25 µmol cDCE (corresponding to an initial concentration of ~5 µmol•L −1 in the aqueous phase). SPME and cDCE depletions were monitored over time. Microbial growth was monitored through protein content in microbial cell pellets obtained from 4 mL of culture. The pellets were further resuspended in 0.5 mL of deionized water adjusted with 0.1 M NaOH and boiled for 1 minute. The protein concentration (mg•L −1 ) was measured according to the Folin-Ciocalteau reagent-based method 66 , the test colorimetric reaction was conducted in the presence of 0.1 M NaOH and quantification was based on bovine serum albumin standard solutions.
Depletion of cDCE in the selected SPME-enriched cultures. The cDCE transformation yield was calculated as the ratio between total cDCE depleted (µmol) and total SPME consumed (mmol) in each vial during the 10 days of cultivation.
Analytical methods used to quantify SPMEs and cDCE. A volume of 0.5 µL of the PCR product was used as the template for another round of PCR, which was performed under the same conditions except that the final reaction volume was 25 µL, with 1 µM of each primer, and the cycle number was decreased to 5-10. The forward and reverse primers used for the second PCR were modified with sequencing adapters and internal barcodes of variable length (5-8 bp) using the TaggiMatrix spreadsheet courtesy of Travis C. Glenn at the University of Georgia (http://www.baddna.org).
The resultant PCR products were purified using AMPure XP Beads (Agencourt, Beckman Coulter, USA). Further amplicon-sample library preparation and sequencing analysis on an Illumina MiSeq instrument were performed at the Core Facility for Nucleic Acid Analysis at the University of Alaska Fairbanks.

Sequencing data analysis.
A pipeline for the analysis of amplicon sequences was adapted from Callahan, et al. 68 using DADA2. The slight modifications made to the DADA2 SOP resulted from the analysis of a mock community, which consisted of 12 bacterial strains and was amplified along with the sample sequences. Briefly, octuplicate sequences for each biological replicate (fastaq files) were filtered according to their quality (parameters truncLen = c(260,260), maxN = 0, maxEE = 2, truncQ = 2), while the primer sequences were trimmed off to the final degenerated position. Upon dereplication, a DADA2-based removal of sequencing errors was performed, followed by the merging of the denoised forward and reverse reads and the removal of chimeric sequences. Based on the results of mock community analysis, sequences differing by one base were clustered together, with the most abundant sequence being regarded as correct; octuplicate sequences were merged, while keeping only those sequences that occurred in at least three of the octuplicates. Finally, taxonomy was assigned to the valid sequences using rdp_train_set_14 69 to create a table of sequence variants. Further sequence analyses were carried out in R project 70 using the phyloseq package 71 , with diversity indexes being calculated on the basis of rarefied data (14,000 valid reads); ordination analysis was carried out on CSS-normalized data 72 , while abundance barcharts were constructed from merged biological replicates. The structural diversity of the community was described by the number of sequence variants observed, Simpson's Index of Diversity (1-D), Simpson's reciprocal index (1/D) and the Shannon index 73,74 . The Bray-Curtis distance-based unconstrained ordination method for NMDS was used to evaluate the ecological distance between the four enriched microbial cultures.
Sequence deposition. MiSeq reads were deposited in the NCBI Short Read Archive under SRA study number SRP090758.