A variety of nucleic acid amplification methods, such as polymerase chain reaction (PCR), nucleic acid sequence-based amplification (NASBA), self-sustained sequence replication (3SR), and rolling circle amplification (RCA) are available. However, LAMP assay is gaining popularity as a point-of-care and other diagnostic applications due to its simplicity and amplification of DNA/RNA with high specificity within 15–60 min under isothermal conditions without the need for thermocycler [11]. Furthermore, positive reaction can be visually determined without agarose gel electrophoresis when using a pH-based and a non-pH-based colorimetric indicators such as phenol red and hydroxynaphthol blue in the reaction mixture, respectively [11, 14]. Since LAMP is a simple, rapid, and sensitive assay, our first goal was to design M. haemolytica species-specific LAMP primers targeting the well-conserved lktD gene of the leukotoxin operon [20]. Purified genomic DNAs from M. haemolytica genotype 2 (strains D153 and D174) and genotype 1 (strain D171) were initially compared with four related Pasteurellaceae species; B. trehalosi, M. glucosida, P. multocida, and H. somni, since they also are opportunistic pathogens that reside as commensals of the upper respiratory tract of cattle [21, 22]. The colorimetric LAMP with lktD primers showed a positive detection (indicated by a color change from pink to yellow) only with M. haemolytica strains and not with the related Pasteurellaceae species (Fig. 1a, lanes 4–7). Although rsmL LAMP produced positive results for M. haemolytica, it also showed a positive result for M. glucosida (Fig. 1b). The observation of ladder-like banding patterns on agarose gels with the positive LAMP samples including M. glucosida (which was amplified with rsmL but not lktD primers) confirmed the colorimetric LAMP results (Fig. 1c, d). These findings suggested that lktD (but not rsmL LAMP primers) can be used to identify M. haemolytica.
Although multiple LAMP assays are available to detect pathogens relevant to food animals, only one study has been conducted so far to detect BRDC bacterial pathogens [19]. In that study, the M. haemolytica LAMP was species-specific and was unable to discriminate serotypes and genotypes [19]. Furthermore, under our experimental conditions, rsmL LAMP primers also amplified M. glucosida genomic DNA suggesting a lack of specificity of rsmL primers for M. haemolytica. The second goal of this study was to develop LAMP primers to discriminate genotypes 1 and 2. We used the adhesin pseudogene B1 and the adhesin G gene, which have been observed in genotype 1 and genotype 2 strains, respectively [9], to design LAMP primers for genotype discrimination. A positive reaction with hydroxynaphthol blue in the colorimetric LAMP assay was indicated by a color change from violet to sky blue. As expected, adhesin pseudogene B1 primers produced positive LAMP results only with a genotype 1 strain (Fig. 2a) while adhesin G primers produced positive LAMP results only with two genotype 2 strains (Fig. 2b). No amplification (as indicated by no color change) was observed with related Pasteurellaceae species (Fig. 2a, b). The observed ladder-like patterns on agarose gels only in lanes with positive LAMP samples further confirmed the colorimetric LAMP findings (Fig. 2d, e). The lack of adhesin gene amplification with H. somni and P. multocida genomic DNA was expected since neither adhesin pseudogene B1 nor adhesin G were detected in the genomes of both species in a recently completed study [23].
To determine how early amplification was completed, a real-time LAMP assay was performed using a multi-purpose LAMP kit supplemented with a LAMP fluorescent dye. Analysis of real-time data revealed that genotype 2 specific LAMP primers showed increased fluorescence signal after ~ 17 cycles (~ 8.5 min) and maximum signal after ~ 35 cycles (17.5 min; red dotted-line, Fig. 2c). However, increased signal for genotype 1 specific LAMP primers was at ~ 47 cycles (~ 23.5 min) and maximum signal was at ~ 80 cycles (~ 40 min; blue dotted-line, Fig. 2c). Although genotype 2 LAMP primer set has both loop forward and loop backward primers, only one loop primer for genotype 1 could be generated. Therefore, the apparent delay in amplification of genotype 1 might be attributed to the lack of one loop primer. Similar observations have been previously reported for loop primers [12].
To further confirm the genotype-specificity of primers, we performed LAMP assay with nine genotype 1 and ten genotype 2 M. haemolytica strains which were previously characterized for adhesin genes [9]. Based on the previous study, all nine genotype 1 strains used in this study were ST2 [9]. Five of the previously untyped genotype 2 strains were serotyped by rapid plate agglutination test in this study for ST1 and ST6. Two of the five strains were identified as ST1 while remaining three were identified as ST6. Representative colorimetric LAMP results of eight strains for each genotype are shown in shown in Additional file 1: Fig. S1. As predicted, genotype 1 primers were specific to genotype 1 strains while genotype 2 primers were specific to genotype 2 (Additional file 1: Fig. S1).
Next, we examined the sensitivity of genotype-specific primers by colorimetric LAMP with the purified genomic DNA from one genotype 1 and two genotype 2 strains. Genotype 1 specific primers were able to detect 100 copies per reaction (Fig. 3a) while genotype 2 specific primers were able to detect 1 copy to 10 copies per reaction (Fig. 3b, c). Agarose gel electrophoresis of LAMP products were consistent with colorimetric LAMP findings (Fig. 3d).
Taken together, these results indicate that colorimetric LAMP assay can be used to rapidly discriminate M. haemolytica genotype 1 and 2. Therefore, genotype-specific LAMP should assist in proper diagnosis of BRDC.