intI 1 gene abundance from septic tanks in Thailand using validated intI 1 primers

Abstract

Antimicrobial resistance is a global crisis, and wastewater treatment, including septic tanks, remains an important source of antimicrobial resistance (AMR) genes. The role of septic tanks in disseminating class 1 integron, and by extension AMR genes, in Thailand, where antibiotic use is unregulated remains understudied. We aimed to monitor gene abundance as a proxy to infer potential AMR from septic tanks in Thailand. We evaluated published intI1 primers due to the lack of consensus on optimal Q-PCR primers and the absence of standardization. Our findings confirmed septic tanks are a source of class 1 integron to the environment. We highlighted the significance of intI1 primer choice, in the context of interpretation of risk associated with AMR spread from septic tanks. We recommend the validated set (F3-R3) for optimal intI1 quantification toward the goal of achieving standardization across studies.

Keywords: AMR; ARGs; Thailand; class 1 integron integrase; decentralized wastewater treatment; intI1; intI1 qPCR primers; septic tanks; wastewater treatment.

Fig 1

Workflow of integrase sub-databases construction for primer evaluation from 922 IntI1 (A) and 2,462 non-IntI1 protein sequences (B). Duplicate IntI1 protein sequence (n = 1) was discarded. Retained protein sequences were compared to the reference IntI1 protein of pVS1 plasmid (AAA25857.1) using NCBI BlastP to identify true intI1 sequences. IntI1 protein sequences with a ≥98% identity to the pVS1 protein sequence were classified as IntI1 protein sequences. Conversely, IntI1 protein sequences with a <98% identity to the pVS1 protein sequence were classified as IntI1-like sequences. Three intI1 gene nucleotide sub-databases (SDB1, SDB2, and SDB3) were finally constructed based on criteria specified in Table 1 and were used to evaluate the coverage of primers. intI1-Like (n = 15) and non-intI1 (n = 1540) sub-databases were used to evaluate the specificity of primers. * Indicates removal of 1 protein sequence (CP006631.1) from the 921 non-duplicate intI1 protein sequence, due to no similarity score to the IntI1 pVS1 protein sequence generated, as a result of low sequence similarity. ^# Indicates the two (WP_058137959.1 and WP_058135314.1) IntI1 protein sequences incorrectly identified as IntI1-like protein sequences by the low similarity score generated by NCBI following alignment to the pSV1 protein sequence due to these sequences being partial length sequences.

Fig 2

Impact of primer choice on the quantification of intI1 gene copies from CST-household, CST-healthcare, and SST-household septic tank wastewater reactors, and three wastewater sample types (influent, effluent, and sludge). Results of the two-way ANOVA analysis showed a statistically significant difference in intI1 gene copies quantified between reactor types and sample types. For each primer set, a boxplot sharing the same letter indicates no statistically significant difference at P-value > 0.05, while the boxplot with different letters indicates a statistically significant difference at P-value < 0.05. A statistically significant difference in intI1 gene abundance between primer sets for the same sample was not observed (P-value > 0.05; see supplementary table s.vi). X icon indicates mean intI1 copy number/ng DNA. The black dot indicates the data outlier.

Fig 3

Detected ASVs abundance in Thai septic tank wastewaters (SST-household, CST-healthcare, and SST-household) by the DF-DR (A), F3-R3 (B), and F7-R7 (C) intI1 primer sets. Generated ASVs coupled with known and unknown intI1 within SDB1 (n = 104), best hit NCBI sequences, and intI1-like sequences were aligned with Mafft, trimmed to only aligned region with no gaps, and phylogenetic tree constructed using the RAxML with 1,000 bootstrap permutations. The number at a node represents a bootstrap value > 50% (from 1,000 permutations). The bootstrap value at node < 50 is not shown. The class 3 integron-integrase (intI3) gene (nucleotide ID: AY219651.1), which on the protein level, shared a 60.74% similarity to the pVS1 protein sequence (AAA25857.1) was used as the outgroup. The color of tree tips indicates the isolated source of sequence/ ASVs generated by the primer set. Heatmap shows log2 fold abundance (mean number of ASVs-DF-DR:5.1955 × 10⁴, F3-R3: 4.6602 × 10⁴ and F7-R7: 3.6684 × 10⁴; Table S7) of detected ASVs within each wastewater sample. CTP3 and CTJ6 samples originated from two independent CST-household reactors. CT-HC sample was from a CST-healthcare tank. ST01 and ST07 are two independent SST-household units. The sampling month and year are indicated by the format month_year (i.e., 06_19 = June 2019). CST, conventional septic tank; SST, solar septic tank.

Fig 4

IntI1 DNA and mRNA transcript quantified from a river water sample by the three selected intI1 primer sets (DF-DR, F3-R3, F7-R7).The reverse transcriptase reaction for each primer set was performed with random hexamers (RH) and gene-specific (GS) primers. In addition, TaqMan assays were carried out with (C) and without the probes (i.e., SYBR Green) (A, B). NT denotes not-tested, and ND denotes non-detected.