key: cord-0770686-4kt1niaw
authors: Warmt, Christian; Yaslanmaz, Ceren; Henkel, Jörg
title: Investigation and validation of labelling loop mediated isothermal amplification (LAMP) products with different nucleotide modifications for various downstream analysis
date: 2022-05-03
journal: Sci Rep
DOI: 10.1038/s41598-022-11320-7
sha: f1546db8db83e49d23be0ae576e5b32e0e748864
doc_id: 770686
cord_uid: 4kt1niaw

Loop mediated isothermal amplification (LAMP) is one of the best known and most popular isothermal amplification methods. It's simplicity and speed make the method particularly suitable for point-of-care diagnostics. Nevertheless, false positive results remain a major drawback. Many (downstream) applications are known for the detection of LAMP amplicons like colorimetric assays, in-situ LAMP or CRISPR-Cas systems. Often, modifications of the LAMP products are necessary for different detection applications such as lateral flow assays. This is usually achieved with pre-modified primer. The aim of this study is to evaluate amplicon labelling with different modified nucleotides such as Cy5-dUTP, biotin-dUTP and aminoallyl-dUTP as an alternative to pre-labelled primers. To realise this, the effects on amplification and labelling efficiency were studied as a function of molecule size and nucleotide amount as well as target concentration. This research shows that diverse labelling of LAMP amplicons can be achieved using different, modified nucleotides during LAMP and that these samples can be analysed by a wide range of downstream applications such as fluorescence spectroscopy, gel electrophoresis, microarrays and lateral flow systems. Furthermore, microarray-based detection and the ability to identify and distinguish false positives were demonstrated as proof of concept.

. LAMP dUTP-labelling strategies. In this study, direct and indirect labelling of LAMP amplicons is shown. (A) Via indirect labelling, 5-aminoallyl-dUTP was introduced into the LAMP products and Cyanine-5 NHS-ester was coupled in a subsequent reaction. (B) Using direct labelling, modifications were incorporated into LAMP products using Cy5-dUTP and biotin-dUTP without additional reactions. (C) Amplicons labelled directly or indirectly via modified nucleotides can be used in various applications like microarray technology, spectroscopic analysis or lateral flow based systems. www.nature.com/scientificreports/ In this study labelling and amplification efficiency using different amounts of x-labelled nucleotides and SARS-CoV-2 cDNA during amplification were compared performing qLAMP and spectroscopic analysis.

Various detection methods were used to show a broad range of possibilities for downstream analysis like fluorescence spectroscopy analysis and microarray technology as well as lateral flow based detection.

Due to a lack of literature on microarray based LAMP detection for the distinction of false positives, this work will be considered as proof of concept study.

For this reason, labelling of Salmonella and E. coli DNA with Cy5-dUTP was performed and LAMP was coupled with a subsequent microarray analysis to enable the identification of false positive results.

Effect of x-dUTP incorporation on LAMP efficiency. To determine the effect on amplification efficiency by using labelled nucleotides during LAMP, SYBR green based quantitative loop mediated isothermal amplification (qLAMP) with various amounts of SARS-CoV-2 cDNA and 5-(3-aminoallyl)-2'-deoxyuridine-5'-triphosphate (Cy5-dUTP) as well as biotin-X-(5-aminoallyl)-dUTP (biotin-dUTP) and 5-aminoallyl-dUTP (AA-dUTP) dilution series was performed. With cDNA amounts of 4.2E + 00 ng per reaction and Cy5-dUTP concentrations ranging from 4.0E-03 mM to 2.0E-01 mM, target DNA has been detected in less than 5 min (Fig. 2) . Sigmoidal like curves could be observed with SYBR green relative fluorescence intensities (rFI) ranging from 1.0 rFI for reactions without Cy5-dUTP and 0.93 rFI using 4.0E-03 mM Cy5-dUTP to 0.15 rFI for the highest Cy5-dUTP concentration (2.0E-01 mM) per reaction ( Fig. 2A) . There is only a slight difference in reaction efficiency using this amount of target DNA, as mirrored by the qLAMP cycle threshold (Ct) ranging from 2.96 min for the positive control (no Cy5-dUTP) and 3.07 min using 4.0E-03 mM Cy5-dUTP to 5.03 min when using 2.0E-01 mM labelled nucleotides. Maximum fluorescence intensity as well as Ct values decreased or increased in the order of labelling molecules per LAMP reaction.

Nevertheless, a decrease in amplification efficiency has been observed when using lower amounts of target DNA. The addition of only 4.2E-02 ng cDNA per LAMP reaction increased the overall Ct-values (Fig. 2B) . While a Ct difference of 2.96 min occurred when high amounts of DNA were used, a Ct-value shift of more than 13.00 min from the lowest to the highest amount of Cy5-dUTP per reaction in diluted samples was observed. A no enzyme control (NEC) was used as negative control with maximum amounts of Cy5-dUTP and showed no amplification curves.

Since Cy5-dUTP is a comparatively large molecule in contrast to unlabelled nucleotides, we investigated the effects on amplification efficiency using various dUTP labels with different molecule sizes. For this AA-dUTP (589.2 g/mol), biotin-dUTP (862.7 g/mol) and Cy5-dUTP (1387.3 g/mol) were used with different amounts ranging from 0.28% to 12.5% (4.0E-03 mM to 2.0-01 mM) compared to the overall x-dUTP + dTTP amounts (Fig. 3) .

In the tested range of x-dUTP endpoint amplification could be observed in less than 22 min, except for the 8.0E−02 mM and 2.0E−01 mM Cy5-dUTP samples. The use of the above mentioned amount of fluorescent-labelled nucleotides results in an increase of the enpoint amplification time to up to 27 min and 35 min, respectively. A clear correlation was found between the molecule size, represented as molecular weight, and the www.nature.com/scientificreports/ required amplification time. While using AA-dUTP, only slight qLAMP Ct-value shifts of less than 3 min from zero percent to 12.5% AA-dUTP per LAMP reaction are detected. This value increases to 4.4 min when using the slightly larger biotin-dUTP and is more than 13 min using Cy5-dUTP.

To investigate the influence of labelling on the amplification under realistic assay conditions, 10E + 04 copies of the SARS-CoV-2 cDNA fragment were amplified with 1.5% x-dUTP/98.5% dTTP each. Successful and complete endpoint amplification was observed for all samples within 25 min (Fig. 3B) . The unlabelled reference sample showed a Ct-value of 15.6 min. Labelling with biotin-dUTP and AA-dUTP also resulted in Ct-values of less than 16.5 min. The use of Cy5-dUTP influenced the amplification time the most, with a Ct-value shift of 3.5 min.

Cy5-dUTP incorporation and -analysis. The rate of Cy5 incorporation into LAMP amplicons was calculated using direct labelling via Cy5-dUTP and an indirect method via AA-dUTP with subsequent NHS-Cy5 labelling. To overcome inconsistent results by uncompleted LAMP reactions an endpoint LAMP (60 min) was performed as previously described. After a preliminary cleaning step, our LAMP products were analysed by fluorescence spectroscopy recording emissions spectra in a 600 nm to 800 nm range with 630 nm excitation (Fig. 4A ). All samples (Cy5 dilution series, no-Cy5 positive control, pure water and no enzyme control) indicated a sharp peak with maximum intensity at 630 nm and normalized fluorescence units (nFU) between Calculation of fluorophores measured in the purified LAMP amplicons was performed with the help of standard calibration line between 7.8E-03 mM and 2.5E-01 mM of free Cy5-dUTP (Fig. 4B ). Using the values for the linear equation, calculation of the concentrations of incorporated, labelled nucleotides per LAMP reaction was possible. By that, the rate of Cy5-dUTP incorporation per 100 unlabelled dTTPs was determined as 0.05 by using 4.0E-03 mM Cy5d-dUTP per LAMP reaction and 1.70 with a ratio of 12.5% labelled dUTP/ 87.5% unlabelled dTTP. Furthermore, an average of 14.4% of all Cy5-dUTP used during LAMP was taken up by the Bst3.0 polymerase, with no correlation to the Cy5-dUTP concentration per LAMP approach (Table 1) .

Additionally, LAMP products were analysed via standard gel electrophoresis using common SYBR green labelling ( Fig. 4C ) and the incorporated Cy5 molecules (Fig. 4D) . In a SYBR green based visualization, all products amplified in an end point LAMP showed gel bands consistent in intensity and band pattern except NEC. Analysing the gel with Tecan LS reloaded microarray scanner (gel tray printed with 3D printing technology), different gel band and fluorescent intensities could be observed. While no-Cy5 positive control and NEC showed no signal, very weak fluorescent bands were seen in the samples with 4.0E-03 mM and 1.0E-02 mM Cy5-dUTP per LAMP reaction. By increasing the amount of labelled nucleotides for LAMP, the increased incorporation into amplicons became visible by increasingly brighter fluorescence bands. www.nature.com/scientificreports/ In addition to the direct Cy5-dUTP labelling, the incorporation rate of small nucleotide modifications has been investigated. This was done by performing LAMP with AA-dUTP and subsequent Cy5-labelling described in the method section and Fig. 1 . As before, the substitution rate was determined by fluorescence measurements. Using only 0.28% modified nucleotides, a substation rate (0.051 dUTP per 100 dTTP) comparable to the direct Cy5 labelling could be calculated. Increasing the AA-dUTP amount leads to slightly but not significantly higher rates. Thus, approximately 1.9 aminoallyl nucleotides out of 100 possible positions were incorporated when LAMP was performed with 2.0E-01 mM AA-dUTP. The mean overall yield of incorporated AA-dUTP of all available is 15.7%.

Microarray analysis with pre-labelled LAMP amplicons. The amplification time for our standard LAMP assay for Salmonella spp. is < 40 min, which is adequate for many applications. Nevertheless, when using 60 min to force false positive results, which is a common drawback of LAMP, false positives in the E. coli spp. control as well as in the no template controls (NTC) were observed (Fig. 5A ). Cy5-dUTP LAMP and subsequent microarray analysis were combined to distinguish Salmonella amplicons and false positives. Prior to microarray hybridization, LAMP products were cleaved with an endonuclease and a restriction enzymes (DNAse and Rsal) and the products verified via gel electrophoresis (Fig. 5B ). After 60 min LAMP with 6 ng genomic DNA per reaction, typical LAMP gel bands of none-cleaved products were detected in Salmonella as well as E. coli and NTC samples. By looking only at the band pattern in the gel, one cannot distinguish the different samples, except that NTC has a slightly different band pattern in some but not all cases.

The treatment of LAMP products with the specific enzyme (RsaI) and unspecific endonuclease (DNase) produced smaller fragments with a majority of less than 100 bp (Fig. 5B ). For these products it was also not possible to distinguish between Salmonella, E. coli and NTC samples. After hybridising treated and untreated samples with Salmonella-specific probes, we could see clear differences in signal intensities in the false colour images (Fig. 5D ). Enzyme treated and untreated controls (NTC and E. coli) as well as treated Salmonella samples showed fluorescence intensities only slightly brighter than the background. On the other hand, untreated Salmonella LAMP amplicons produced signal intensities clearly different from background and controls.

This could also be observed from the graphical evaluation (Fig. 5C ). For the untreated Salmonella sample, the InvA-gen and LAMP sequence specific probes invA_1990 and invA_1994 showed signal intensities of more than 6000 FU and 4000 FU respectively. In addition, lower intensities of almost 1000 FU were detected for invA_1925, but no signals with invA_1933. Controls (E. coli and NTC) produced only 10%-19% signal intensities using invA_1990 and invA_1994 probes.

Control probes (OXA48 and VIMP1) as well as buffer spots showed no fluorescence.

Modifying LAMP products for a broad range of diverse applications is a commonly used technique and is regularly achieved by the incorporation of pre-modified primer sets. Due to the complex primer set of up to six primers, the design and the evaluation of LAMP primers is not as simple as the design of PCR primers and can lead to false positive results if not carefully validated. Furthermore, in the methods described so far, the primers must be additionally prepared with appropriate molecules to enable their use, e.g. for LFA assays or spectroscopic analyses. In this study, we describe another promising method for labelling amplicons without using previously labelled primers. Table 1 . Yield of Cy5-dUTP incorporation. Depicted are the calculations for the x-dUTP incorporation during LAMP. First column shows the x-dUTP dilution series used in the different LAMP reactions followed by the column for the overall ratio of labelled dUTPs to unlabelled dTTPs in the reaction that is equal to the maximum resulting theoretical incorporation of labelled dUTP per 100 unlabelled dTTPs. The labelling columns showing the true incorporation rate per 100 dTTPs followed by the overall rate of incorporate x-dUTP out of all x-dUTP inserted in LAMP. Using direct labelling, labelling efficiency for large molecules like Cy5-dUTP were determined. Performing indirect labelling with subsequent fluorescent labelling, the incorporation yield using small molecules like AA-dUTP has been investigated (compare Fig. 1 ). www.nature.com/scientificreports/ Modifications of LAMP products by using dUTP instead of dTTP for usage in uracil DNA glycosylase (UDG-LAMP) assays have been shown previously 30, 31 indicating that LAMP enzymes like Bst2.0 and Bst3.0 are able to handle this nucleotide modification without causing serious amplification problems. Only a handful of publications are known in which additional modifications like digoxigenin-dUTP (DIG-dUTP) 32 as well as fluorescent 29 or biotin labelled nucleotides 23 have been used.

In this study, various modified uracil nucleotides (x-dUTP) for the labelling of LAMP products were tested and validated. For this purpose, 5-(3-aminoallyl)-2'-deoxyuridine-5'-triphosphate (Cy5-dUTP), biotin-X-(5-aminoallyl)-dUTP (biotin-dUTP) and 5-aminoallyl-dUTP (AA-dUTP) represent different molecules with different sizes and properties. Cy5-dUTP allows direct labelling and the fluorescence signal can be used for amplicon detection directly after LAMP. Biotin-dUTP and AA-dUTP require at least one further reaction step. For example, Biotin-labelled amplicons can be used for immobilization via a biotin-streptavidin interaction in lateral flow assays (LFA). The aminoallyl-modification enables the downstream coupling of a wide range of other NHS-functionalised molecules for a broad spectrum of applications.

This study investigated whether the different modifications can be incorporated into the amplicons during the LAMP reaction. Furthermore, the effects on the LAMP reaction at different concentrations of the labelling molecules and the ratio of the incorporated labels were investigated.

To achieve this, two different labelling strategies were used. First, direct labelling with Cy5-dUTP and biotin-dUTP was performed without downstream reaction. Secondly, indirect labelling was carried out by incorporating functional groups (aminoallyl group) followed by chemical coupling of the target molecule (in this case NHS-Cy5). www.nature.com/scientificreports/ Performing different detection methods such as qLAMP or spectroscopic and fluorescent gel analysis, it was not only possible to examine the influence of different molecule sizes but to show a broad range of downstream applications.

A commercially available LAMP kit, as described in the Methods and Materials section, was used for all experiments. To this, 7.5% DMSO was added as a preventive treatment to reduce false positives. This method has been described previously 33 and showed the best results in our preliminary work. According to our knowledge, the DMSO has no influence on the labelling and was not explicitly added for this purpose.

This study shows that the introduction of different modifications for labelling LAMP products is possible. Small functional groups as well as larger molecules such as biotin and fluorescent dyes can be integrated into the amplicons. As shown in Fig. 2 and Fig. 3 , amplification efficiency highly depends on the type of nucleotide modification, size and concentration. For the different modifications, a range from 0.28% x-dUTP/99.72% dTTP to 12.5% x-dUTP/87.5% dTTP was teste successfully.

When using large amounts of target DNA, in our example 4.2E + 00 ng of a 466 bp SARS-CoV-2 N-gene cDNA fragment, only Cy5-dUTP shows a slight impact of ΔCt = 2.96 min when using 12.5% Cy5-dUTP/87.5% dTTP compared to an unmodified amplification. The influence increases with the decrease of the target DNA. A 100-fold reduction to 4.2E-02 ng cDNA increased the ∆Ct to nearly 14 min using 12.5% Cy5 dUTP but only 30-60 s with 1.5% Cy5-dUTP. The effect of the Ct-value shift decreases also when using smaller nucleotide labels like biotin (4.4 min at 12.5%) or aminoallyl-group (2.9 min at 12.5%).

Cy5-dUTP-labelled LAMP products were successfully analysed by fluorescence spectrometry after amplification. It was shown that the fluorescence intensity of the amplicons increases with increasing concentration of the labelled nucleotides during LAMP. Fluorescence intensity increased 45-fold when the amount of Cy5-dUTP per LAMP reaction was increased from 4.0E-03 mM (0.28%) to 2.0E-01 mM (12.5%). On the basis of the data available, calculation of a Cy5-dUTP/dTTP substitution rate was possible of up to 1.7 per 100 dTTP when using 12.5% dUTP and about 0.2-0.5 per 100 dTTP when using 1.5%-2.8% dUTP during the LAMP reaction. Using only 0.3% fluorescence modified nucleotides, the labelling rate is less than 0.05 per 100 dTTP. Thus fluorescence intensity can be easily adjusted by using different amounts of labelling molecules during the LAMP process. Nevertheless, the average substitution rate of incorporate Cy5-dUTP out of all Cy5-dUTP available in LAMP is 14.4%.

It was also investigated whether the efficiency of labelling can be increased by using the indirect method. As shown in Fig. 1A , aminoallyl-modified dUTPs were incorporated into the amplicons for this purpose. Subsequently, the fluorescent dye NHS-Cy5 was coupled to the functionalised LAMP products and fluorescence intensity was measured as before. This showed that the use of smaller molecules (e.g. AA-dUTP compared to Cy5-dUTP) leads to better amplification efficiency, but does not seem to have a significant effect on labelling efficiency. The average total yield of incorporated AA-dUTP is 15.7% compared to 14.4% using Cy5-dUTP. This suggests that the activity of the LAMP enzyme is highly affected as a function of molecule size, but the rate of substitution is only affected slightly. Therefore, direct labelling is more reasonable as long as the labelling nucleotides with the desired modification are available. This eliminates post-processing steps such as coupling reactions and purification of the products. The only advantage of using smaller modifications is the speed of the LAMP assay as long as no subsequent coupling reaction is required. Nevertheless, the indirect method is a viable alternative if the desired modifications cannot be incorporated directly during LAMP.

In Agarwal, S. et al., 2022 we used this fact to overcome problems with rapid SARS-CoV-2 detection. While fluorescein isothiocyanate labelling via FITC-dUTP in combination with a biotin-labelled primer seems to work for 30-min LAMP by Zhang, C. et al. 2021, we could not detect LAMP products after 10-min amplification time with our own assay. Changing the system to biotin-dUTP and FITC-labelled primer solved the problem 23 .

The use of 1.5% modified nucleotides under realistic assay conditions (Fig. 3B ) resulted in only a marginal slowing of the amplification rate of less than 1 min when using Biotin-dUTP and AA-dUTP and just 3.5 min with Cy5-dUTP. In this case, 10E + 04 copies of SARS-CoV-2 cDNA fragments were used as starting material. This amount corresponds to the viral load in the sample of a mildly to moderately infected person 34 . These results show that sufficient labelling of amplicons can be achieved with both small molecules (AA-dUTP) and large fluorophores without significant loss of time. The use of less starting material (10E + 03 and 10E + 02 copies of cDNA) also resulted in successful amplification for all modifications used without false positives or false negatives. However, wider variations in Ct-values for both unlabelled and labelled amplicons made it impossible to determine the exact Ct-value shifts for this range (data not shown).

As a compromise between fluorescent intensity and amplification speed we recommend the use of only 1% to 2% labelled nucleotides when using large molecules like fluorophores and up to 5% with molecule sizes like biotin-dUTP.

Following this recommendation, it was possible to detect LAMP products not only via fluorescence spectroscopy and fluorescence gel electrophoresis but also by microarray analysis and 1.4% Cy5-dUTP per LAMP.

Microarray detection is a common and valuable tool for the analysis of PCR and isothermal amplification products 25, 35, 36 . Recently, we were able to present nucleotide-based labelling during recombinase polymerase amplification (RPA) followed by microarray as a very advantageous method 26 . In this study, we have successfully transferred this technique in a most intriguing way to detect and distinguish LAMP products.

Salmonella LAMP with 1.4% Cy5-dUTP and subsequent microarray detection was performed. As a proof of principle analysis this study investigated three main criteria. The first question related to the feasibility of detecting labelled amplicons using microarray technology in general, which was proved in this study. The second part was to investigate whether it is possible to discriminate false positive results. To this purpose, a 60-min LAMP was performed with false-positive no-template controls (NTC) and false-positive E. coli controls beside Salmonella samples. While the band pattern in the agarose gel (Fig. 5A) indicates different LAMP products in NTC and E. coli samples, no differences could be detected between E. coli and Salmonella samples. Using microarray technology and specific probes, we were finally able to distinguish the latter. The third aim was to investigate www.nature.com/scientificreports/ whether digestion of the LAMP products with restriction enzymes or nucleases is required to hybridize them successfully. For microarray detection, the usage of small DNA fragments is recommended to reduce difficulties in hybridization. Digestion of large DNA sequences can improve signal intensity 36, 37 . Because LAMP products can extend to very long lengths due to the amplification mechanism, Salmonella, E. coli and NTC amplicons were digested with nonspecific DNase and specific RsaI enzymes. While digestion of amplicons with both enzymes was detectable by gel electrophoresis, only untreated samples could be detected and distinguished by microarray. These results indicate that LAMP amplification can be combined directly with microarray detection to analyse and distinguish amplicons. Moreover, it can be performed without purification by simply adding hybridisation buffer after the LAMP reaction and incubation on an array. Thus, this application has the potential to be more specific than the commonly used lateral flow or colorimetric downstream detection methods. In this work, it was shown that the labelling of LAMP products using x-dUTP with 1-2% of labelled nucleotides produces excellent signals depending on the detection method used. We did not detect any significant loss of amplification or analytical performance compared to the previously published labelling methods. Further investigations on this, especially in direct comparison with primer-based labels, are currently the subject of further studies. However, results to date indicate that at least equivalent analytical performance is achieved with lateral flow-based, microarray-based and spectroscopic analyses. In addition, this method offers the possibility of optimising sensitivity by increasing the incorporation rates of labelling molecules, which is not possible when using single-labelled primers. In combination with suitable detection methods such as microarray technology, it should be possible to perform multiplex LAMP analyses without the need to use a large number of labelled primers.

In summary, this study shows that custom and flexible labelling of LAMP amplicons is possible using different modified nucleotides incorporated directly during the amplification process.

Thus, an easy-to-use and flexible alternative to the commonly used primer-based labelling was demonstrated, which is able to adjust the labelling efficiency more flexibly depending on the experimental question. This allows easy adaptation to different downstream detection methods such as fluorescence spectroscopy, gel electrophoresis, microarrays, lateral flow systems or in-situ LAMP without the need for a pre-labelled primer.

LAMP reaction and labelling reagents. 25 µl LAMP with Bst3.0 DNA Polymerase and reaction buffer (New England Biolabs; M0374S) was performed according to manufactures instructions (using a Bio-Rad CFX Real-Time PCR System) accept to the following changes: i) for all assays dimethyl sulfoxide (DMSO) was used with a final concentration of 7.5% (v/v) and the volume of DNA template per reaction was 1 µl; ii) for qLAMP assay 1 µl of 50 × SYBR green I (Invitrogen; S7563) was added; iii) aminoallyl-dUTP-XX-Cy5 (Jena Bioscience; NU-803-XX-CY5-L), biotin-11-dUTP (Jena Bioscience; NU-803-BIOX-L) and 5-aminoallyl-dUTP (Biutium; #40,020) were used for labelling of LAMP products with final concentrations ranging from 4.0E-03 mM to 2.0E-01 mM (0.28% to 12.5%) per reaction. The Amount of nuclease free water was adjusted to fit 25 µl reaction volume. DNA template concentration varied due to experimental settings (see results section). Primer and sample sequences are listed in Table 2 .

Depending on the experimental question, amplification times have been set to 20-60 min at 65 °C followed by an enzyme denaturation step of 5 min at 80 °C. For samples that were to be cleaved by restriction enzymes or nucleases, an additional heating step for 10 min at 90 °C was performed.

LAMP post-preparation. LAMP products were purified with Mag-Bind® Total Pure NGS Kit (Omega Bio-Tek; M1378-01) according to the manufactures instructions and eluted in 30 µl ddH 2 O for subsequent gel electrophoretic and spectroscopic analysis. For spectroscopic analysis only, the fluorescently labelled DNA was purified by washing five times with 70% ethanol (v/v). Aminoallyl-products were purified twice. Once immediately after LAMP reaction to remove all amplification reagents (5 times ethanol wash) with a final elution step in 30 µl 0,1 M NaCO 3 -buffer (pH 8,4) as reaction medium for Cy5-NHS coupling. For this purpose, Cyanine-5 NHS-ester (Lumiprobe; 43,020) was pre-dissolved in DMSO and water was added to a final concentration of 10 mM. 2.5 µl was added to 22.5 µl LAMP product in NaCO 3 -buffer, mixed gently and incubated at 23 °C for 3 h in the dark. Afterwards the second purification step was performed with 10 times ethanol wash for complete removal of unbound fluorophores, ending with an elution step in 30 µl ddH 2 O.

Purification of microarray samples not treated with restriction enzymes was not performed. Digestion of LAMP products using RsaI was performed by using 20 µl purified LAMP product (approx. 2 µg DNA), 2 µl enzyme (10 U/µl; New England Biolabs; R0167S) and 5 µl associated buffer (New England Biolabs; B6004S). Reaction mixture was filled up with ddH 2 O to final volume of 25 µl.

After digestion at 37 °C for 15 min, reaction was stopped by adding 10 µl of corresponding gel loading dye (New England Biolabs; B7024A).

Digestion of LAMP product using RQ1 DNase was performed by using 20 µl purified LAMP product (approx. Templates and Primer. For the determination of labelling efficiency including the qLAMP, fluorescent spectroscopy and gel electrophoretic experiments a 466 bp cDNA fragment from SARS-CoV-2 N-gene (provided by Robert Koch Institute (RKI)) was used as well as LAMP primer designed with PrimerExplorer V5 (Table 2) . Primers were designed based on the genome reference sequence MN908947.3 (GenBank, Table 2 ). The cDNA was amplified from the genomic RNA of the respective viruses in preparatory work. The extracted RNA was provided by the RKI. 38 .

Probes were designed and literature primer were compared based on the genome reference sequence M90846.1 available in GenBank ( Table 2) .

All primers were synthesized and provided by metabion AG.

Gel electrophoresis. For gel electrophoretic analysis, 3 µl DNA Gel Loading Dye (6 x) (Thermo Fisher Scientific; R0611) were added to the purified (and digested) samples except the RsaI approach. For band visualization, a 2% agarose gel was prepared in 1 × TAE buffer (50 × TAE buffer; PanReac AppliChem; A1691) containing 4 µl peqGreen (VWR; 732-3196) per 100 ml gel. 12-15 µl of the purified and prepared samples were used for electrophoresis at 120 V for 1 h before detection via BioDoc Analyzer (Biometra) using a Canon EOS 1100D. GeneRuler Low Range DNA Ladder (Thermo Fisher Scientific; SM1191) and peqGOLD DNA ladder (VWR; were used for calculation of fragment length. For the detection of incorporated Cy5-dUTP the DNA gel was scanned with a Tecan LS reloaded scanner using a 3D-printed gel tray. Table 2 . List of primer, probes and target sequences. LAMP sequences (F3 to B3 primer region) are underlined and bold letter mark the product from FIP to BIP primer. Complementary regions for microarray probes are bold underlined and recognition sequence for RsaI restriction enzyme is depicted in low case letters. For Salmonella invA gene, only the relevant sequence region is shown. InvA gene bases not shown are represented by […] . LAMP primer for Salmonella spp. were first published by Youn et al. 38 . B3/F3 bases in square brackets do not match our Salmonella sequence, but the LAMP still works very well. * amino-C6-T 13 ; ** microarray negative control. ATC AGC GAA ATG CAC CCC GCATTA CGT TTG GTG GAC CCT CAG ATT CAA CTG GCA GTA ACC AGA ATG GAG  AAC GCA GTG GGG CGC GAT CAA AAC AAC GTC GGC CCC AAG GTT TAC CCA ATA ATA CTG CGT CTT GGT TCA CCG CTC TCA  CTC AAC ATG GCA AGG AAG ACC TTA AAT TCC CTC GAG GAC AAG GCG TTC CAA TTAA CAC CAA TAG CAG TCC AGA TGA CCA  AAT TGGCT ACT ACC GAA GAG CTA CCA GAC GAA TTC GTG GTG GTG ACG GTA AAA TGA AAG ATC TCA GTC CAA GAT GGT ATT  TCT ACT ACC TAG GAA CTG GGC CAG AAG CTG GAC TTC CCT ATG GTG CTA ACA AAG ACG GCA TCA TAT GGG TTG CAA CTG  AGG GAG CCT TGA ATA CAC CAA AAG ATC ACA TTG GCA CCC GCA ATC CTG CTA ACA ATG CTG CAA TCG TGC TAC TAC AAG CAT GAA ATG GCA GAA CAG CGT CGT ACT ATT GAA AAG CTG TCT TAA TTT AAT ATT AAC AGG ATA  CCT ATA GTG CTG CTT TCT CTA CTT AAC AGT GCT CGT TTA CGA CCT GAA TTA CTG ATT CTGgtacTAA TGG TGA TGA TCA TTT  CTATG TTC GTC ATT CCA TTA CCT ACC TAT CTG GTT GAT TTC CTG ATC GCA CTG AAT ATCgtacTGG CGA TAT TGG TGT TTA  TGG GGT CGT TCT ACA TTG ACA GAA TCC TCA GTT TTT CAA CGT TTC CTG CGgtacTGT TAA TTA CCA CGC TCT TTC GTC TGG  CAT TAT CGA TCAgtacCAG TCG TCT TAT CTT GAT TGA AGC CGA TGC CGG TGA AAT TAT CGC CAC GTT CGG GCA ATT CGT TAT  TGG CGA TAG CCT GGC GGT www.nature.com/scientificreports/ Fluorescent spectroscopy. The efficiency of fluorescence labeling with Cy5-dUTP was analysed with the PerkinElmer LS55 using FL Winlab software (PerkinElmer) and FLUOstar Omega (BMG Labtech). Using LS55 fluorescence spectrometer, purified LAMP products were diluted in a ratio of 1:16 (v/v) with a final volume of 96 µl and analysed using the following parameters: (i) 630 nm excitation and emission spectrum ranging from 600 to 800 nm; (ii) Emslit/Exslit = 10 nm; (iii) scan speed = 100 nm/min. All measurements were conducted as triplicates.

Using FLUOstar Omega, 30 µl purified samples were analysed without dilution using 620 nm excitation and 680 nm emission filter.

For the standard calibration line, free Cy5-dUTP was taken at a dilution range of 7.8E-03 mM to 2.5E-01 mM.

Microarray analysis. Microarray analysis was performed using the hybridization protocol published in Warmt et al. 26 . For this purpose, the corresponding probes were spotted onto 3D-epoxy glass slides (PolyAn GmbH) using a sciFLEXARRAYER SX (Scienion GmbH). After blocking with ethanolamine and subsequent washing steps, hybridization took place at 52 °C for 90 min. Subsequently, several washing steps were performed before the slide was analysed on the GenePix 4300 microarray scanner.

Probe sequences are listed in Table 2 .

Concentration measurements. Concentration measurements were conducted using Nanodrop 2000 (Thermo Fisher Scientific) with 1.5 µl purified LAMP products.

The data that support the findings of this study are available from the corresponding author [C.W.] upon reasonable request. The reference sequences analysed and used for primer and probe design during the current study are available in GenBank under accession number MN908947.3 (SARS-CoV-2) and CP074639.1 (Salmonella enterica).

Loop-mediated isothermal amplification of DNA

LAMP diagnostics at the point-of-care: Emerging trends and perspectives for the developer community

Accelerated reaction by loop-mediated isothermal amplification using loop primers

Loop-mediated isothermal amplification for direct detection of Mycobacterium tuberculosis complex, M-avium, and M-intracellulare in sputum samples

Real-time reverse transcription loop-mediated isothermal amplification for rapid detection of West Nile virus

Rapid detection and typing of influenza A and B by loop-mediated isothermal amplification: Comparison with immunochromatography and virus isolation

African trypanosomiasis: Sensitive and rapid detection of the sub-genus Trypanozoon by loop-mediated isothermal amplification (LAMP) of parasite DNA

Demonstration of a quantitative triplex LAMP assay with an improved probe-based readout for the detection of MRSA

A LAMP (loop-mediated isothermal amplification) test for rapid identification of Khapra beetle (Trogoderma granarium)

Poly-adenine-coupled LAMP barcoding to detect apple scar skin viroid

A simple, affordable, rapid, stabilized, colorimetric, versatile RT-LAMP assay to detect SARS-CoV-2

PCR is changing clinical diagnostics: PCR outperforms traditional culture-based diagnostic procedures that often miss bacteria when they fail to grow

LAMP-on-a-chip: Revising microfluidic platforms for loop-mediated DNA amplification

Loop-mediated isothermal amplification (LAMP): A rapid, sensitive, specific, and cost-effective point-of-care test for coronaviruses in the context of COVID-19 pandemic

Pullulan reduces the non-specific amplification of loop-mediated isothermal amplification (LAMP)

Simultaneous elimination of carryover contamination and detection of DNA with uracil-DNA-glycosylase-supplemented loop-mediated isothermal amplification (UDG-LAMP)

Loop-Mediated Isothermal Amplification for Detection of African Trypanosomes

Real-time quantitative and ion-metal indicator LAMP-based assays for rapid detection of Sclerotinia sclerotiorum

Rapid and simple colorimetric loop-mediated isothermal amplification (LAMP) assay for the detection of Bovine alphaherpesvirus 1

Method for colorimetric detection of double-stranded nucleic acid using leuco triphenylmethane dyes

LAMP assays for the simple and rapid detection of clinically important urinary pathogens including the detection of resistance to 3rd generation cephalosporins

Development of a low-cost copro-LAMP assay for simultaneous copro-detection of Toxocara canis and Toxocara cati

Lateral flow-based nucleic acid detection of SARS-CoV-2 using enzymatic incorporation of biotin-labeled dUTP for POCT use

RT-LAMP CRISPR-Cas12/13-Based SARS-CoV-2 Detection Methods

Using Cy5-dUTP labelling of RPA-amplicons with downstream microarray analysis for the detection of antibiotic resistance genes

Zika virus lateral flow assays using reverse transcription-loop-mediated isothermal amplification

An integrated lateral flow assay for effective DNA amplification and detection at the point of care

Rapid one-pot detection of SARS-CoV-2 based on a lateral flow assay in clinical samples

An advanced uracil DNA glycosylase-supplemented loop-mediated isothermal amplification (UDG-LAMP) technique used in the sensitive and specific detection of Cryptosporidium parvum, Cryptosporidium hominis, and Cryptosporidium meleagridis in AIDS patients

Development of uracil-DNA-glycosylase-supplemented loopmediated isothermal amplification coupled with nanogold probe (UDG-LAMP-AuNP) for specific detection of Pseudomonas aeruginosa

In situ loop-mediated isothermal amplification (LAMP) for identification of Plasmodium species in widerange thin blood smears

Two methods for increased specificity and sensitivity in loop-mediated isothermal amplification

Virological assessment of hospitalized patients with COVID-2019

Multiplex isothermal solid-phase recombinase polymerase amplification for the specific and fast DNA-based detection of three bacterial pathogens

DNA microarrays: Types, Applications and their future

Direct detection and genotyping of Klebsiella pneumoniae carbapenemases from urine by use of a new DNA microarray test

Application of loop-mediated isothermal amplification with propidium monoazide treatment to detect live Salmonella in chicken carcasses

C.W. was responsible for the concept, coordination and monitoring of the experimental setup and the study, the spectroscopic analysis as well as the review and evaluation of all data and results and the writing of the manuscript. C.Y was responsible for the microarray based lab experiments and preparatory works for spectroscopic analysis as well as reviewing the manuscript. J.H. supervised the study and was responsible for reviewing the data and the manuscript.

Open Access funding enabled and organized by Projekt DEAL.

The authors declare no competing interests.

Correspondence and requests for materials should be addressed to C.W. or J.H.Reprints and permissions information is available at www.nature.com/reprints.Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.