PCE, trichloroethylene (TCE), cis-1,2-dichloroethylene (cDCE), 1,2-DCA (all ≥99%) were purchased from Macklin Biochemical Co., Ltd (Shanghai, China). Vinyl chloride (VC) and ethene (both ≥99%) were purchased from Dalian Special Gases Co., Ltd (Dalian, China). CuCl₂·2H₂O, and ZnCl₂ were purchased from Sigma-Aldrich (St. Louis, MO, USA). CdCl₂ was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China). All other chemicals used in this study were analytical grade or of higher purity.

Medium preparation and anaerobic cultivation were performed following estabolished protocols (Löffler et al., 2005; Yang et al., 2017a; Yang et al., 2017b). Briefly, reduced, bicarbonate-buffered mineral salts medium was boiling under an atmosphere of N₂ to remove dissolved oxygen, cooled down to room temperature and then dispensed into serum bottles flushed with N₂/CO₂ (80/20, v/v). Sediment samples were collected from the Xi River, Shenyang, Liaoning Province, China (41°44′41″N, 123°17′35″E) as described (Jiang et al., 2022). Microcosms were constructed inside an anoxic chamber (Coy Laboratory Inc., MI, USA) filled with N₂/H₂ (97/3, v/v). An aliquot of 2 mL sediment sludge was pipetted into 160 mL glass serum bottles prefilled with 100 mL of medium mentioned above as described (Yang et al., 2017b; Yang et al., 2020). Bottles were sealed with butyl rubber stoppers (Fushiyuan rubber and plastic products factory, Shenzhen, Guangdong, China) and crimped with aluminum caps (Hongpu Instrument Technology, Ningbo, Zhejiang, China). Lactate (10 mM) was provided as the carbon source and electron donor. A dose of 5 μL neat PCE (ca. 0.31 mM or 51.41 mg/L aqueous phase concentration) or 5 μL neat 1,2-DCA (ca. 0.62 mM or 61.36 mg/L aqueous phase concentration) was added as the electron acceptor. All bottles were amended with Wolin vitamin mix (Wolin et al., 1963). A heavy metal ion (e.g., Cu²⁺, Cd²⁺, Zn²⁺) was added into each bottle from a stock solution. For each heavy metal ion, three different concentrations (5, 10, and 50 mg/L) were examined. Following the complete dechlorination of PCE or 1,2-DCA to ethene, 1 mL culture suspensison was transferred into fresh medium following the same procedures described above. Additional dechlorinating microcosms and enrichment cultures that did not receive the amenedment of a heavy metal ion were established as controls. The bottles were incubated statically in the dark at 30°C. Microcosms and enrichment cultures were established in duplicate bottles.

Cells were harvested from 5 mL culture suspension via vacuum filtration onto 25 mm diameter 0.22 μm pore-size membrane filters (Merck Millipore Ltd, Darmstadt, Germany). Genomic DNA was extracted from the filters using the Soil DNA Kit (Tiangen Biotech, Beijing, China) following the manufacturer’s instructions. DNA concentrations were determined using a Qubit 3.0 fluorometer (Invitrogen, Carlsbad, CA, USA). Amplicon sequencing was performed by Suzhou Genewiz Biotechnology (Suzhou, Jiangsu, China). For 16S rRNA gene amplicon sequencing, the primer set Pro341F (5’-CCTACGGRRBGCASCAGKVRVGAAT-3’) and Pro806R (5’-GGACTACNVGGGTWTCTAATCC-3’) modified by Genewiz for relatively conserved regions within the V3 and V4 hypervariable regions were used to generate amplicons (Fu et al., 2016). The quality and quantity of DNA libraries were evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and a Qubit 2.0 Fluorometer, respectively. DNA libraries were multiplexed and loaded on an Illumina MiSeq instrument (Illumina, San Diego, CA, USA) following the manufacturer’s instructions. Sequencing was performed using a 2x250 paired-end (PE) configuration and image analysis and base calling were conducted using the MiSeq Control Software (MCS) embedded in the MiSeq instrument. The raw data were processed using Cutadapt (v1.9.1), VSEARCH (v1.9.6), and QIIME (v1.9.1) with default parameters to obtain clean reads (Caporaso et al., 2010; Bhute et al., 2016). Following quality control, sequences were grouped into operational taxonomic units (OTUs) using VSEARCH and QIIME at a 97% sequence identity. All OTUs were assigned to the lowest possible taxonomic rank using RDP classifier 2.2 and Silva_132 16S rRNA database (Wang et al., 2007; Edgar, 2010). Number of the obtained sequences in each sample was between 58,306 and 182,628, and a total of 686 OTUs were clustered from all the samples. Amplicon sequencing reads were deposited into Figshare repository with the following access address: https://doi.org/10.6084/m9.figshare.19212618.v1.

Ethene and chlorinated compounds were analyzed using an Agilent 7890B gas chromatography equipped with an Agilent 7697A automatic headspace sampler, a flame ionization detector (FID) (method detection limit ~ 0.2μM) and an Agilent DB-624 capillary column (60 m length x 0.32 mm inner diameter x 1.8 μm film thickness) as described (Yang et al., 2017b). Oven temperature was initially held at 60°C for 2 min, increased to 200°C at 25°C/min, and held at 200°C for 1 min. Inlet and FID temperatures were set at 200°C and 300°C, respectively.

Principal component analysis (PCA) is a widely used statistical technique for dimension reduction and can be used for resolving differences among samples with distinct treatments. PCA was performed using the PCA plug-in (version 1.50) in OriginPro 2021 (OriginLab Corp. Northampton, MA, USA). Heatmaps of microbial communities were plotted using the ggplot2 packages (Wickham, 2011) embedded in R language (version 4.1.1) (R Core Team, 2020).

PCE and 1,2-DCA dechlorinating cultures were enriched from urban river sediments with lactate as the electron donor and carbon source. The enrichment cultures maintained stepwise PCE-to-ethene dechlorination and 1,2-DCA-to-ethene dihaloelimination after two consecutive transfers (Figure 1A). In the PCE-dechlorinating enrichment cultures, the initial 46.40 ± 0.72 μmol PCE was dechlorinated to stoichiometric amounts of ethene within 30 days via hydrogenolysis with TCE, cDCE, and VC being sequentially produced. Geobacter and Dehalococcoides were the dominant OHRB genera in the second transferred enrichment cultures with relative abundances of 11.12% and 3.42%, respectively. Geobacter lovleyi strain SZ, isolated from Su-Zi Creek sediment of Korea in 2005, and Geobacter lovleyi strain LYY, isolated from black-odorous urban river sediment of China, are capable of PCE-to-cDCE dechlorination (Sung et al., 2006; Amos et al., 2007; Liang et al., 2021). This information suggested that the Geobacter population in the PCE-dechlorinating enrichment was responsible for the reductive dechlorination of PCE to cDCE, which was futher dechlorinated to ethene by Dehalococcoides.

In the 1,2-DCA-dechlorinating enrichment cultures, the intial 72.86 ± 3.09 μmol 1,2-DCA was dihaloeliminated to stoichiometric amounts of ethene within 30 days. Dehalobacter was the most abundant OHRB genus in the second transfer cultures with a relative abundance of 24.52% (Figure 1C), indicating that Dehalobacter may be responsible for dihaloelimination of 1,2-DCA. Dehalobacter sp. strain WL is able to dechlorinate 1,2-DCA to ethene using a novel reductive dehalogenase (RDase), which shares a 92% similarity to the dihaloeliminating DcaA of Desulfitobacterium dichloroeliminans strain DCA1 (Grostern and Edwards, 2006; van der Zaan et al., 2009). To date, the ability to perform dihaloelimination only has been identified in four organohalide-respiring genera including Dehalobacter, Desulfitobacterium, Geobacter and Dehalogenimonas (De Wildeman et al., 2003; Grostern and Edwards, 2009; Yan et al., 2009; Jiang et al., 2022).

To investigate the effects of heavy metal ions on dechlorination processes, sediment microcosms were set up with urban river sediments to dechlorinate PCE or 1,2-DCA at the presence of elevated concentration (i.e., 5, 10 or 50 mg/L) of a heavy metal ion (i.e., Zn²⁺, Cu²⁺ or Cd²⁺). In control cultures (i.e., without the addition of heavy metal ions), complete PCE-to-ethene dechlorination was achieved in microcosms and the transferred enrichment cultures within one month (Figure 1A). Complete dechlorination of PCE to ethene was also accomplished within one month in sediment microcosms amended with a heavy metal ion (i.e., Zn²⁺, Cu²⁺ or Cd²⁺) at a concentration up to 50 mg/L (Figure 2A), suggesting that the tolerance of OHRB to heavy metal ions can be greatly enhanced with protective solid matrix (e.g., minerals). Subsequent transfers of the PCE-dechlorinating enrichment cultures at the presence of relatively low concentrations (i.e., 5 or 10 mg/L) of a heavy metal ion were successful (Figure 2B), indicating that the enrichment cultures inherited the ability to tolerate the toxicity of Zn²⁺, Cu²⁺ and Cd²⁺ in low concentrations. By comparison, PCE-to-ethene dechlorination could not be sustained and stalled activity was observed in the first and second transfer cultures when exposed to 50 mg/L Zn²⁺ or 50 mg/L Cu²⁺ during a 80 days’ incubation (Figure 2A). PCE was completely dechlorinated to ethene in the transferred enrichment cultures exposed to 50 mg/L Zn²⁺ after incubated for about 200 days; nevertheless, complete PCE dechlorination did not occur in transferred enrichment cultures at the presence of 50 mg/L Cu²⁺ after incubated for one and half years (data not shown). In the first transfer, reduced PCE dechlorination rate (i.e., 40 days for complete dechlorination) was observed in enrichment cultures at the presence of 5 or 10 mg/L Zn²⁺. When the concentration of Zn²⁺ was increased to 50 mg/L, 7.04 ± 1.97 μmol TCE, negligible cDCE and VC were detected in the first transfer on day 68 (Figure 2A). The inhibitory effects of 50 mg/L Cu²⁺ on the dechlorination of PCE were similar to Zn²⁺ (Figure 2A). Unexpectedly, up to 50 mg/L Cd²⁺ did not affect PCE-to-ethene dechlorination activity, and complete PCE dechlorination to stoichiometric amounts of ethene was accomplished within 40 days (Figures 2A, B).

Without the addition of a heavy metal ion, sediment microcosms and transferred enrichment cultures completely dechlorinated the initial 72.86 ± 3.09 μmol 1,2-DCA to stoichiometric amounts of ethene within 25 days (Figure 1A). In sediment microcosms amended with up to 50 mg/L of a heavy metal ion (e.g., Cd²⁺, Zn²⁺ or Cu²⁺), 1,2-DCA could also be dihaloeliminated to ethene within 30 days (Figure 3A), indicating that these heavy metal ions at a concentration less than 50 mg/L had no apparent inhibitory effects on the reductive dechlorination of 1,2-DCA. Nevertheless, 1,2-DCA-to-ethene dechlorination could not be sustained and was completely lost in the first transfer cultures exposed to 50 mg/L Zn²⁺ and the second transfer cultures exposed to 50 mg/L Cu²⁺ (Figure 3A). Unlike the PCE-dechlorinating enrichment cultures, 1,2-DCA was not dechlorinated to ethene after prolonged incubation in the transfer cultures amended with 50 mg/L Zn²⁺ or Cu²⁺ (data not shown). By comparison, up to 10 mg/L Zn²⁺ or Cu²⁺ had little impacts on the dechlorination of 1,2-DCA in the second transfer cultures, and 72.86 ± 3.09 μmol 1,2-DCA was completely dechlorinated to ethene within 30 days (Figure 3B). Surprisingly, Cd²⁺ had little inhibitory effects on the dechlorination of 1,2-DCA as well, and 1,2-DCA was completely dechlorinated to ethene within 30 days in microcosms and transfer cultures at the presence of as high as 50 mg/L Cd²⁺ (Figures 3A, B).

To evaluate effects of heavy metal ions on the dechlorinating communities, 16S rRNA gene amplicon sequencing was applied to samples collected from PCE- or 1,2-DCA-dechlorinating microcosms and transfer cultures exposed to different concentrations of heavy metal ions. Proteobacteria, Bacteroidetes, Firmicutes and Chloroflexi were the dominant phyla in sediment microcosms and enrichment cultures. The abundances of Bacteroidetes and Firmicutes increased when PCE or 1,2-DCA was provided (Figure 1B). Geobacter and Dehalococcoides were the main OHRB genera in the second transfer of PCE-dechlorinating cultures where a heavy metal ion was not added. Of note, the relative abundances of Geobacter and Dehalococcoides were 0.58% and undetectable, respectively, in the river sediment sample (Figure 1C). Dehalobacter was the most abundant OHRB genus in the second transfer of 1,2-DCA-dechlorinating enrichment cultures without the addition of a heavy metal ion;however, Dehalobacter was not detected (i.e., below the detection limit) in the river sediment sample (Figure 1C). Rikenellaceae, Lentimicrobium, Sedimentibacter, and Youngiibacter were likely the major fermenters in PCE-dechlorinating microcosms that produced acetate and H₂ to supply Dehalococcoides and Geobacter. By comparison, Clostridium, Lentimicrobium, Rikenellaceae, Smithella and Acetobacteroides were the most abundant fermenters potentially producing acetate and H₂ for Dehalobacter in the 1,2-DCA-dechlorinating enrichment cultures.

Principal component analysis (PCA) demonstrated that the concentrations and types of heavy metal ions and the types of organohalides were the predominant factors shaping the community structures (Figure 4). Geobacter, Dehalococcoides, Dehalogenimonas, and Desulfovibrio were clustered under the PCE-dechlorinating condition, while Dehalobacter was enriched under the 1,2-DCA-dechlorinating condition (Figure 4A). Moreover, community structures of the enrichment cultures received an addition of 50 mg/L Cu²⁺ or 50 mg/L Zn²⁺ differed greatly from the enrichment cultures that did not receive the addition of a heavy metal ion and other enrichment cultures, especially the enrichment cultures amended with Cd²⁺ (Figures 4B, C).

We also observed that the addition of a heavy metal ion (i.e., Zn²⁺, Cu²⁺, Cd²⁺) likely promoted the enrichment of OHRB, especially the obligate OHRBs (Figures 5, 6). In the second transfer of PCE-dechlorinating enrichment cultures amended with a heavy metal ion, the relative abundances of Dehalobacter increased from 0.02% in the control cultures to 4.5% in the enrichment cultures amended with 5 mg/L Zn²⁺, and 17.2% in the enrichment cultures amended with 10 mg/L Zn²⁺. Dehalococcoides, Dehalogenimonas, and Desulfovibrio were also enriched with the addition of a heavy metal ion. For instance, the relative abundances of Dehalogenimonas, of which certain member is capable of dechlorinating TCE to ethene (Yang et al., 2017b), increased to 48.1% in the enrichment cultures exposed to 50 mg/L Zn²⁺ from 0.1% in the control cultures. Likewise, the relative abundances of Dehalococcoides was 11.2% in the enrichment cultures exposed to 10 mg/L Cu²⁺, compared with 3.5% in control cultures and 5.7% in cultures amended with 5 mg/L Cu²⁺. In contrast, decreasing abundances of Geobacter was found in the enrichment cultures exposed to a heavy metal ion during consective transfers. The relative abundances of Geobacter decreased from 11.1% in the control cultures to 6.2% in the enrichment cultures exposed to 50 mg/L Cd²⁺. The fermenter Lentimicrobium was seemingly tolerant to heavy metal ions, and the relative abundances were 4.5%, 8.9% and 15.7% in control cultures, the second transfer cultures exposed to 10 mg/L Zn²⁺ and 10 mg/L Cd²⁺, respectively.

In the 1,2-DCA-dechlorinating cultures, Dehalococcoides and Desulfovibrio were greatly enriched in Cu²⁺ amended enrichment cultures. The relative abundance of Dehalococcoides in the second transfer of 1,2-DCA-dechlorinating enrichment cultures amended with 50 mg/L Cu²⁺ was 5.7% compared with 0.3% in the control cultures. The relative abundance of Desulfovibrio in the second transfer of 1,2-DCA-dechlorinating enrichment cultures amended with 10 mg/L Cu²⁺ was 8.7%, compared with 0.2% in the control cultures. Similar to the PCE-dechlorinating enrichment cultures, Lentimicrobium was enriched in 1,2-DCA-dechlorinating enrichment cultures exposed to a heavy metal ion, and the relative abundances were 14.0%, 18.3%, 18.5% and 20.8% in control cultures, the second transfer of enrichment cultures exposed to 10 mg/L Zn²⁺, 10 mg/L Cu²⁺ and 10 mg/L Cd²⁺, respectively.