Rapid visualization and detection of Staphylococcus aureus based on loop-mediated isothermal amplification

Staphylococcus aureus is a common clinical bacterial pathogen that can cause a diverse range of infections. The establishment of a rapid and reliable assay for the early diagnosis and detection of S. aureus is of great significance. In this study, we developed a closed-tube loop-mediated isothermal amplification (LAMP) assay for the visual detection of S. aureus using the colorimetric indicator hydroxy naphthol blue (HNB). The LAMP reaction was optimized by adjusting the amplification temperature, the concentrations of Mg2+, dNTP, and HNB, and the incubation time. In the optimized reaction system, the specificity of LAMP for S. aureus was 100%. The results established that this method accurately identified S. aureus, with no cross-reactivity with 14 non-S. aureus strains. The limit of detection (LOD) of LAMP was 8 copies/reaction of purified plasmid DNA or 400 colony-forming units/reaction of S. aureus. Compared with conventional PCR, LAMP lowered the LOD by tenfold. Finally, 220 clinically isolated strains of S. aureus and 149 non-S. aureus strains were used to evaluate the diagnostic efficacy of LAMP (test accuracy, 99.46%). The findings indicated that LAMP is a reliable test for S. aureus and could be a promising tool for the rapid diagnosis of S. aureus infections.


Introduction
Staphylococcus aureus is a common human pathogen that secretes a variety of virulence factors with significant pathogenicity (Otto 2014). S. aureus can cause a wide range of clinical infections, from localized skin and soft tissue infections to life-threatening necrotizing pneumonia (Oliveira et al. 2018). It is also the main cause of bacteremia and osteoarticular, pleuropulmonary, and device-related infections (Tong et al. 2015). S. aureus bacteremia is highly prevalent and extremely hard to treat, leading to increased mortality and the burden of additional medical costs (Jin et al. 2021;Naber 2009). Therefore, the development of a reliable and rapid test to accurately detect S. aureus presents an urgent clinical need.
The conventional detection of S. aureus is based on microbial culturing, including bacterial colony morphology, biochemical reactions, and antimicrobial susceptibility testing (Sheet et al. 2016). Although this method has good sensitivity and reliability, it has some obvious limitations. For example, some bacterial species grow only in specific nutrient matrices and their identification requires various selective media (Lazcka et al. 2007). Moreover, the culturing of strains is a labor-intensive and time-consuming process, taking approximately 2-5 days to produce the desired results (Zhou et al. 2021). Consequently, many patients may miss the optimal period of treatment. Moreover, the improper use of antibiotics may lead to bacterial resistance (Paule et al. 2005). In recent decades, various molecular diagnostic methods have been devised and applied for the classification and identification of pathogenic microorganisms (Mwaigwisya et al. 2015). Compared with microbiological culturing, molecular diagnostic technology has higher sensitivity, better reproducibility, and a shorter time required to produce results. For example, PCR technology 1 3 209 Page 2 of 9 as an alternative to microbiological culturing is popular in the field of pathogenic microbial diagnosis because of its great amplification performance (Florio et al. 2018). Several studies have reported on PCR assays for the detection of S. aureus in clinical samples with a high diagnostic accuracy rate (Brakstad et al. 1992;Wang et al. 2019). However, a PCR assay requires expensive experimental instruments, reagents, and skilled personnel, which may not be readily available in many resource-poor areas. Therefore, a costeffective, simple, and reliable assay for the identification of S. aureus should be developed as an alternative routine laboratory test.
Loop-mediated isothermal amplification (LAMP) is a viable alternative to PCR in the molecular diagnosis of bacterial pathogens. LAMP technology utilizes four primers to identify the target sequence with high selectivity and good amplification specificity (Notomi et al. 2000). It can amplify a limited initial number of DNA copies to one million copies in one hour at isothermal conditions and does not require pre-denaturation of the template (Notomi et al. 2015;Nagamine et al. 2001). Recently, LAMP technology has been applied to the detection of several pathogenic bacteria, parasites, viruses, and diseases (Wong et al. 2018;Mori and Notomi 2009;Fang et al. 2010). However, the conventional LAMP technique analyzes amplification results by agarose gel electrophoresis or turbidimetry, which often require extended workflow times or specialized equipment (Mori et al. 2001). Direct visual detection of turbidity to determine whether amplification has occurred is the simplest and most cost-effective method, but it is difficult to observe the precipitate even under ideal conditions, limiting the widespread use of LAMP in clinical laboratories (Mori et al. 2004;Miyamoto et al. 2015). Hydroxy naphthol blue (HNB) is a metal ion indicator, which can visualize DNA amplification by detecting the decrease of Mg 2+ ion concentration during the LAMP reaction process (Goto et al. 2009). HNB is easy to use and the colors of negative and positive reactions are significantly different, which enables the LAMP reaction to achieve a clear and rapid visual detection.

Bacterial strains and genomic DNA template preparation
A total of 15 bacterial strains were used to analyze the specificity of the LAMP reaction, as listed in Table 1. A total of 369 bacterial strains were isolated from patients at Traditional Chinese Medicine Hospital of Sichuan and Chengdu Fifth People's Hospital from January to May 2021 (Table S1). These strains were identified at the species level by the VITEK 2 compact system (Biomerieux, France) which was used as the reference diagnostic results for diagnostic efficacy evaluation of the LAMP and the PCR assays. For enrichment culture, all strains were inoculated in tryptic soy broth (TSB) liquid medium and incubated at 37 °C with shaking at 200 rpm for 12 h. Ezup Column Bacteria Genomic DNA Purification Kit (Sangon, China) was used to extract the genome DNA following the manufacturer's protocols, and then stored at -20 ℃ for subsequent experiments. For the tenfold serial dilution of S. aureus (ATCC 25923), used for sensitivity assessment, we extracted its genomic DNA using 5 × Hotshot + Tween reagent (Brewster et al. 2013).

Screening of target genes and primer design
We initially screened for genes specific to S. aureus by local BLAST and online BLAST was used to further identify the target gene, according to previously published procedures . Following the results of local BLAST and online BLAST screening, a highly conservative specific sequence in the S. aureus MJ163 chromosome (GenBank Accession Number, CP038229.1) was selected as the target gene in this study. Then, a set of LAMP primers (Table 2) targeting 6 distinct regions on the target gene were designed using Primer Explorer V5 (https:// prime rexpl orer. jp/ lampv5/ index. html), including forward inner primer (FIP, consisting of the F2 sequence at its 3′ end and F1c region at its 5′ end), backward inner primer (BIP, consisting of the B2 sequence at its 3′ end and B1c region at its 5′ end), the outer primers  S1). The primer design and screening process were carried out according to the primer designing manual (http:// prime rexpl orer. jp/e/ v5_ manual/ index. html). The specificity of primers was validated by BLASTN against the GenBank nucleotide database.

Construction and verification of positive control plasmids
The positive control plasmids were constructed by the following steps. First, the DNA fragment of the target gene was amplified by PCR with primers F3 and B3 using the S. aureus genome DNA as a template. The target fragment was then inserted into the pEASY®-T3 Cloning Vector (TransGen, China) and transferred to Trans1-T1 Phage Resistant Chemically Competent cells (TransGen, China). The transformed cells were then incubated in Luria-Bertani liquid medium for enrichment at 37 °C with shaking at 200 rpm for 12 h in the presence of ampicillin. Finally, positive control plasmids were extracted from enriched cells with the TIAN pure Mini Plasmid Kit (Tiangen, China) and stored at − 20 °C for subsequent experiments. All operating procedures were carried out according to the manufacturer's manuscript. The sequences of positive control plasmids were verified by gene sequencing and aligned with the sequences available in GenBank (Fig. S2).

Optimizations for the LAMP assay
The optimization of the LAMP reaction was performed on a G1000 gene amplification instrument (Bioer Technology, China) with purified positive plasmids (8 × 10 5 copies/μL) as templates. Nuclease-free water (TransGen, China) was used as a template for the LAMP blank control. The reaction parameters were sequentially optimized for temperatures in the range of 61-65 ℃, Mg 2+ concentrations between 2.0 and 10.0 mM, dNTP in the range of 0.6-1.4 mM, incubation times between 15 and 65 min, and HNB (Aladdin, China) between 60 and 180 μΜ. Effective amplification resulted in a color change from violet to sky blue. However, the color of the blank control reaction remained violet. Agarose gel electrophoresis was applied for the determination and verification of the amplification of fragments of the correct sizes. The positive amplification of LAMP showed typical ladder bands, while no bands were visible in the negative control. Images were captured by SmartGel 140 (Sage Creation Science, China). All experiments were repeated three times.

Specificity of the LAMP assay
The specificity of the LAMP assay was assessed by detecting genomic DNA extracted from 15 strains of bacterial based on the established optimal reaction system. In addition, a LAMP reaction mixture containing the genomic DNA of non-S. aureus strains was used as a negative control. For the blank control, nuclease-free water was used without any genomic DNA. The amplification results indicated by the color change of HNB dye were verified by agarose gel electrophoresis. All experiments were repeated three times.

Sensitivity of the LAMP assay
Two different templates, the positive control plasmid and genomic DNA of S. aureus, were used to evaluate the limit of detection (LOD) of LAMP reaction under optimized conditions. The extracted positive plasmids, at a concentration of 30.52 ng/µL, were mixed with Tris-EDTA buffer (20 mmol/L Tris-HCl and 2 mmol/L EDTA) and diluted to 8 × 10 8 copies/µL as a working solution, which was then serially diluted tenfold to a working concentration (8 × 10 8 -8 × 10 0 copies/µL). To obtain a specific DNA copy number per µl from the DNA stock solution, we referred to the DNA Copy Number and Dilution Calculator tool (https:// www. therm ofish er. cn/ cn/ zh/ home/ brands/ thermo-scien tific/ molec ular-biolo gy/ molecular-biology-learning-center/ molecular-biology-resource-library/thermo-scientific-webtools/dna-copy-number-calculator.html). The S. aureus bacterial suspensions (4 × 10 8 CFU/mL) was serially diluted tenfold in TSB liquid medium to obtain a series of bacterial solution at various concentrations (4 × 10 8 -4 × 10 0 CFU/ mL). These bacterial suspensions were directly lysed by the 5 × Hotshot + Tween reagent and the resulting solutions were used as the template for LAMP analysis. The LOD of LAMP was calculated based on the minimum mean concentration of the positive plasmid and S. aureus. All experiments were repeated three times.

Application of LAMP in clinical strain samples
The diagnostic efficacy of the LAMP assay was determined using the 369 clinical isolates. These samples were inoculated into TSB liquid medium for enrichment and the genomic DNA was extracted for LAMP experiments. PCR was also performed with specific primers F3 and B3. Finally,  TGG CTT AAT AGA TTA ATT GTT G  B3  ACG CAC TCC TTT TCC AAA T  FIP  ACG GAA TGT AAG CGT AAG AAACG-CGT GCT AAT GTT TTT AAA TGG  ATT C  BIP  AGA AAG TCT AAC GCT TAA AAC TGC  T-GAA TTG CTG CTT GCA TGA C we compared the results of the LAMP assay with those of the PCR in terms of specificity, sensitivity, and test accuracy. The calculation formula were as follows: Specificity = TN/ (TN + FP) × 100%; Sensitivity = TP/(TP + FN) × 100%; Test accuracy = (TP + TN)/(TP + FP + TN + FN) × 100% (TP, true positive; FP, false positive; TN, true negative; FN, false negative).

Optimal reaction system for LAMP
The critical parameters of the LAMP reaction, including amplification temperature, the concentration of Mg 2+ and dNTP, and incubation time, were optimized to achieve optimal amplification efficiency. As shown in Fig. 1a, LAMP electrophoresis bands were brighter at 62 °C than at other amplification temperatures, with no significant differences at 61, 63, 64, or 65 °C. Therefore, 62 °C was selected as the optimal reaction amplification temperature for subsequent experiments. For the optimization of Mg 2+ concentrations at 62 °C, the best amplification efficiency was observed at 8.0 mM Mg 2+ (Fig. 1b). The maximum amount of amplification products was obtained at 1.2 mM dNTP (Fig. 1c). The amplification product was first detected when the reaction was performed for 35 min (Fig. 1d). For optimal amplification, an incubation time of 45 min was subsequently used as a LAMP reaction parameter. Based on the above reaction system, 120 µM HNB was selected in subsequent experiments to achieve optimal visual detection (Fig. 1e).
In summary, the optimized LAMP system used in the present study contained 2.5 µL 10 × isothermal amplifica- 1.2 mM dNTP, 0.8 M betaine, 120 µM HNB, 0.2 µM each of primers F3 and B3, 1.6 µM each of primers FIP and BIP, DNA template and nuclease-free water. The reaction mixture was incubated at 62℃ for 45 min and then heated at 80 ℃ for 2 min to terminate the reaction.

Validation of the specificity of LAMP
A significant color change in the LAMP assay from violet to sky blue was only observed for the S. aureus strain, while the 14 non-S. aureus strains retained their original violet color (Fig. 2a). Agarose gel electrophoresis further confirmed that only the strain of S. aureus showed typical ladder-like pattern bands, while the others showed no bands (Fig. 2b). These results were consistent with the PCR assay using the primers F3 and B3 (Fig. 2c). These consistent results indicated that the LAMP assay that we developed specifically detects S. aureus and has no significant cross-reactivity with other species.

Limit of detection of LAMP
Significant color changes from violet to sky blue were observed in the tenfold serial dilutions of the positive plasmid templates ranging from 8 × 10 8 to 8 × 10 0 copies/reaction when the LAMP assay was performed at 62 °C for 45 min (Fig. 3a). The amplification results were further confirmed on 2% agarose gel electrophoresis (Fig. 3c). However, the LOD of conventional PCR was 8 × 10 1 copies/reaction for the positive plasmid (Fig. 3e). To further explore the sensitivity of the LAMP assay, a suspension of S. aureus was diluted to 4 × 10 8 , 4 × 10 7 , 4 × 10 6 , 4 × 10 5 , 4 × 10 4 , 4 × 10 3 , 4 × 10 2 , 4 × 10 1 , and 4 × 10 0 CFU/mL, respectively. Genomic DNA was extracted from these diluted samples and directly used for LAMP and PCR analysis. The results showed that at a dilution of 4 × 10 3 CFU/reaction S. aureus could be detected by PCR (Fig. 3f), while S. aureus could be detected by LAMP at a dilution of 4 × 10 2 CFU/reaction (Fig. 3b, d). Therefore, the value of LOD of the LAMP assay was tenfold lower than that of PCR. Consequently, for the detection of S. aureus the LOD of the LAMP assay was superior to that of the PCR assay.

Diagnostic efficacy of LAMP
Next, we analyzed 369 clinical isolates to determine the diagnostic efficacy of the LAMP assay. Among 149 non-S. aureus strains, one strain of Klebsiella pneumoniae exhibited non-specific amplification in the PCR reaction (Fig. 4c). Conversely, none of the non-S. aureus strains were effectively amplified in LAMP (Fig. 4a, b). Of the 220 S. aureus-positive samples, 2 strains were deemed negative by LAMP and PCR. Therefore, compared to the reference diagnostic results, the sensitivity, specificity and test accuracy of LAMP assay were 99.09%, 100.00% and 99.46%, respectively. The results of testing clinical strains revealed that LAMP is a reliable and accurate assay for the detection of S. aureus.

Discussion
S. aureus is a common clinical pathogen that causes a variety of serious hospital and community-acquired infections (Holland et al. 2014). Rapid and accurate diagnosis of S. aureus is a key component of treatment, prevention, and interruption of transmission. The currently existing diagnostic methods have limitations as they are either time-consuming (microbial culturing) or expensive (PCR). LAMP is a simple and cost-effective molecular diagnostic method with distinct advantages for the rapid diagnosis of pathogens.
In this study, we successfully developed, optimized, and validated a molecular biology diagnostic LAMP assay for the detection of S. aureus. The critical parameters of the LAMP reaction were optimized to achieve optimal amplification and performance by visual inspection. The optimized LAMP assay can accurately differentiate S. aureus from non-S. aureus strains, and sensitively detect trace amounts of S. aureus.
At present, a variety of colorimetric indicators have been developed for LAMP reaction endpoint detection. Although these indicators are reliable and convenient, they have some limitations. Calcein needs to bind the ionic form of manganese to work, but manganese may inhibit the LAMP reaction and reduce the sensitivity of the assay (Tanner et al. 2015). The fluorescent intercalating dye SYBR Green I is expensive, highly toxic, and requires additional testing equipment (Fischbach et al. 2015). Leuco crystal violet (LCV) requires the preparation of proprietary LCV immobilized tubes, which are complicated to operate (Miyamoto et al. 2015). In contrast, HNB is a cheap and stable synthetic dye. It indicates the result of the LAMP reaction by detecting the consumption of free magnesium ions in the reaction system, which is reliable and effective for LAMP. A multi-observer study showed that HNB as an indicator was superior to the other methods (Wastling et al. 2010). The use of the colorimetric indicator HNB makes the judgment of the LAMP amplification results easier, eliminating the need for tedious electrophoresis operations and greatly avoiding cross-contamination caused by aerosols (Nie 2005). Therefore, HNB was selected as the best dye for this study.
We compared the sensitivity of the LAMP assay with that of the PCR assay and found that the LOD of the LAMP was tenfold lower than PCR, both for the positive plasmid and S. aureus. Several related studies have also confirmed that the sensitivity of LAMP assay was equivalent or higher compared to PCR (Choopara et al. 2021;Maeda et al. 2005;Prusty et al. 2016). In addition, LAMP technology is more convenient, time-saving, and cost-saving. The entire detection process of LAMP, including template preparation (approximately 40 min), isothermal amplification (45 min), and result determination (approximately 2 min), can be completed within 90 min and the cost of analysis is also lower than that of PCR (Zheng et al. 2018). More importantly, LAMP testing is easy to operate, which can reduce labor costs. Finally, LAMP is often described as a less demanding test than PCR in terms of DNA purification. For example, bacterial colonies are added directly to the reaction mixture for the LAMP reaction, or centrifuged pellet samples are boiled in the presence of 1% Triton X-100 and the released DNA can be used directly for the LAMP reaction without purification (Sowmya et al. 2012;Yan et al. 2017).
The diagnostic efficacy of LAMP analysis is shown in Table 3. A Klebsiella pneumoniae strain, isolated from the sputum of a patient with chronic obstructive pulmonary disease, showed a non-specific electrophoresis band by PCR amplification. The electrophoresis results showed that the amplified band size was over 250 bp, while the target fragment band size for S. aureus was 219 bp (Fig. 4c). One possible explanation is that the genome of this strain has some base fragments homologous to the target gene in S. aureus. However, due to the specific amplification mechanism, the LAMP reaction did not produce a false-positive result for this sample. This indicates that the specificity of the LAMP assay (100%) is superior to that of the PCR (99.33%). In other related studies, the specificity of both detection methods for bacterial pathogens was consistent (Thakur et al. 2018;Zhao et al. 2013 were not identified by LAMP and PCR assays, one case isolated from a sputum sample of a patient with cerebral hemorrhage, and the other isolated from the blood of a shock patient. We speculate that this result may be related to a genetic mutation of the strains themselves. Furthermore, the test accuracy of LAMP is higher than that of PCR (99.46% vs. 99.19%), indicating that the accuracy and authenticity of the test results are superior to PCR. Another point worth mentioning is that although the LOD of the LAMP assay was lower than that of the PCR assay, the sensitivity of the PCR assay is equivalent to that of the LAMP assay in clinical strain detection, which may be related to the low number of clinical strains we analyzed. Our study provides a rapid detection protocol for S. aureus. However, for point-of-care testing, further research is still needed. If LAMP technology is integrated into a microfluidic system, DNA extraction, isothermal amplification, and colorimetric analysis could be completed in one step, which will have greater potential for clinical applications (Liu et al. 2020;Wan et al. 2019;Zhang et al. 2019).

Conclusions
In summary, we successfully developed a LAMP assay for the rapid detection of S. aureus. In particular, the colorimetric indicator HNB was used to indicate the LAMP test result, and the positive LAMP reactions were easily identified by the naked eye under ambient light. Application of LAMP for rapid detection of clinical isolates gave a 99.46% test accuracy. LAMP is simpler, faster, and easier to perform than conventional PCR and culture microbiology. Therefore, the present research shows that LAMP can be an effective tool for the immediate diagnosis of S. aureus infection.