In total, 20 multidrug-resistant bacteria without the mcr-1 gene, as determined using the existing PCR method, were used as negative controls (Liu et al., 2016). Two isolates carrying the mcr-1 gene, Escherichia coli 12452 and E. coli 12536, were screened by our laboratory in 2016 and used as positive controls. The information on these strains is shown in Table 1. Genomic DNA from bacteria was extracted using a one-step microbial DNA extraction method based on Chelex-100 (de Lamballerie et al., 1992). In addition, in order to assess the sensitivity of the LAMP assay, genomic DNA was also extracted from E. coli 12452 using a Wizard genomic DNA purification kit (Promega, Madison, WI, USA) according to the manufacturer's instructions. Purified DNA was then prepared by serial 10-fold dilutions to give concentrations ranging from 20 ng/μL to 0.02 pg/μL (Liu et al., 2012).

The sequence of the mcr-1 gene was downloaded from the GenBank database (accession number LC114018.1), and the gene-specific primers were then designed using the Primer Explorer (version 5) software (http://primerexplorer.jp/lampv5e/index.html). First, the forward inner primer (FIP), backward inner primer (BIP), outer forward primer (F3), and outer backward primer (B3) were designed. Then, additional loop primers (LF and LB) were designed to accelerate the amplification reaction (Notomi et al., 2000). All primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China).

Fifty milligrams of fresh feces from healthy individuals was suspended in 200 μL distilled water by vigorous shaking for 5 min. An amount of 10⁸ bacteria (corresponding to 0.2 OD₆₀₀) or dilution was resuspended in 50 μL of 0.85% NaCl and spiked in 200 μL of human fecal suspension (Bontron et al., 2016). The total DNA was extracted from spiked stool samples with a QIAamp DNA Stool Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. DNA was recovered in 100 μL distilled water, and the LAMP assay was performed with 2 μL DNA template.

The LAMP assay was performed in 25 μL reaction mixtures containing the following components: 20 mM Tris-HCl (pH 8.8), 10 mM KCl, 10 mM (NH4)₂SO₄, 8 mM MgSO₄, 0.8 M betaine, 0.1% Tween-20, 1.4 mM of each dNTP, and 8 U Bst DNA polymerase. The amounts of primers needed for one reaction were 40 pmol for FIP and BIP, 20 pmol for LF and LB, and 5 pmol for F3 and B3. Finally, an appropriate amount of genomic DNA template was added to the reaction tube (Liu et al., 2012). The mixture was covered with 25 μL wax and incubated for 60 min in a dry bath incubator with a controlled temperature. Temperature was optimized by testing at 60–67°C.

LAMP assay amplification products were detected using two different methods: turbidimetry and visual detection. For monitoring of turbidity, real-time amplification by the LAMP assay was monitored using a Loopamp real-time turbidimeter (LA320-C; Eiken Chemical Co. Ltd., Tokyo, Japan). For visual detection, manganese ion (Mn²⁺) and calcein, a fluorescent metal indicator, were added to the reaction mixture, which appeared orange because calcein was quenched by Mn²⁺. For a positive reaction, Mn²⁺ formed a magnesium pyrophosphate precipitate (an insoluble salt) during amplification, and the color of the reaction system changed from orange to green. For a negative reaction, there was no color change because amplification did not occur (Mori et al., 2001).

PCR was performed to compare assay sensitivities for detection of the mcr-1 gene using the previously reported CLR5-F and CLR5-R primers (Liu et al., 2016). PCR tubes were set up using 12.5 μL PCR MasterMix reagents (Tiangen Biotech Co., Ltd., Beijing, China), 1 μM CLR5-F and CLR5-R primers (0.4 μM), and 1 μL DNA template in a final volume of 25 μL. The thermal cycler parameters were 3 min at 95°C; 30 cycles of 30 s at 95°C, 30 s at 55°C, and 30 s at 72°C; followed by 5 min at 72°C. The PCR-amplified products were analyzed by 1% agarose gel electrophoresis and stained with GelRed nucleic acid gel stain (Biotium, Hayward, CA, USA). Images were documented using a Bio-Rad Gel Doc EQ imaging system (Bio-Rad, Hercules, CA, USA).

All volunteers provided written, informed consent to participate in this study, which was reviewed and approved by the ethics committee of the Academy of Military Medical Sciences, China. All experiments were performed in accordance with relevant guidelines and regulations.

Five sets of primers for detection of the mcr-1 gene were designed. According to the primer parameters and variants of mcr-1 gene mutations (Thanh et al., 2016), we excluded the fourth and fifth sets of primers and evaluated the other three sets of primers. Under the same reaction conditions, three turbidity curves were observed, indicating that all three primers had the ability to amplify the target fragment. Through this analysis, we found that the third set of primers allowed for completion of the amplification in the shortest time (Figure 1). Primers in the third primer set (Table 2) were chosen as the final primers for mcr-1 gene detection by LAMP. For more information on the five sets of primers, see the Supplementary Materials.

Various reaction temperatures ranging from 60 to 67°C were employed at 1°C intervals to optimize the amplification. As shown in Figure 2, LAMP assay amplification occurred in the range of 60–67°C, and shorter reaction times were observed at 65, 66, and 67°C compared with those at other temperatures. We chose 66°C as the optimum temperature due to the higher peak and stronger enzyme activity in the LAMP assay (the optimum temperature for enzymatic activity of the Bst DNA polymerase was 65°C). Finally, we chose 66°C as the reaction temperature for later experiments.

Next, we analyzed the specificity of LAMP detection of the mcr-1 gene. As shown in Figure 3, only when E. coli 12452 and E. coli 12536 were used as the templates, both turbidity and visual detection appeared positive. Importantly, 20 multidrug-resistant bacteria without the mcr-1 gene, including the blank control, tested negative, indicating that the LAMP assay was highly specific for the mcr-1 gene.

To determine the sensitivity of primers in LAMP detection of the mcr-1 gene, purified genomic DNA was extracted from E. coli 12452 and subjected to serial 10-fold dilutions. PCR was also carried out with the same amount of purified DNA template. As shown in Figure 4, the six reaction curves of real-time turbidity indicated that six serial 10-fold dilutions of the template were detected (20 ng/μL to 0.2 pg/μL), consistent with results of the visual method (positive reactions turned green, whereas negative reactions remained orange). The detection limit of the LAMP assay for the mcr-1 gene was 0.2 pg/μL, which was 10 times more sensitive than conventional PCR.

Because the mcr-1 gene is often detected in human and cattle feces, LAMP assays were applied to human stools spiked with different amounts of mcr-1-positive bacteria. PCR was also carried out as a control. The results showed that the detection limit of the LAMP assay for human stools was 10³ bacteria, which was also 10 times more sensitive than conventional PCR. Thus, LAMP assays can be efficiently applied to detection of the mcr-1 gene in fecal samples.

A LAMP-based surveillance method for detection of the mcr-1 gene in clinical isolates was conducted in a local hospital in Beijing, China. In total, 556 multidrug-resistant Enterobacteriaceae were collected in 2016 from patients with suspected infections; these bacteria had been screened for the mcr-1 gene by PCR using the above primers (Liu et al., 2016).

Subsequently, the results of LAMP assays showed that seven mcr-1-positive isolates, including six E. coli and one Klebsiella pneumoniae, were detected from 556 bacteria. These results were consistent with the results of the PCR assay. Finally, these mcr-1-positive isolates were validated again by PCR-based sequencing. Thus, these findings demonstrated that LAMP assays were well suited for large-scale clinical screening.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.