Perspective Chapter: Recent Progressions on the Colorimetric Diagnosis of SARS-CoV-2 by Loop-Mediated Isothermal Amplification (LAMP) Assay

Abstract

A simple, fast, and accurate diagnosis of SARS-CoV-2 is of great importance for the patient’s isolation, treatment, and the control of the COVID-19 pandemic. Although RT-qPCR is accepted as the gold standard, studies to improve fast, simple, and more reliable diagnostic methods are continuing. Colorimetric reverse transcription loop-mediated isothermal amplification (RT-LAMP) is a method that allows visual detection of SARS-CoV-2 without needing expensive fluorescence readers. However, the performance of the assay depends on some factors, such as selection of a target gene (i.e., N, RdRp, S, E, M), primer design, the dye used for visual observation—neutral red, calcein, cresol red, or phenol red—and the reaction conditions such as the buffer pH, reaction temperature, and enzyme concentration. In the last 2 years, plenty of research has been conducted to obtain the best performance. In this chapter, the recent progressions on colorimetric RT-LAMP assay for the diagnosis of SARS-CoV-2 are comprehensively elucidated.

1. Introduction

For over 2 years from its emergence, the coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has infected over 500 million cases and 6.2 million deaths worldwide as of May 2022 [1]. This highly contagious disease spreads through the respiratory tract via sneezing, coughing, and talking and causes a range of symptoms from mild, like coughing, to severe, like pneumonia. Most patients are asymptomatic and do not require hospitalization; however, high-risk groups—elders over 65 years old, patients with obesity, impaired immune system, and chronic diseases—may develop serious complications that lead to septic shock, multi-organ failure, and eventually, death. However, asymptomatic patients are 75% more likely to spread the disease to others compared to symptomatic cases, which hinders the control of this pandemic [2, 3, 4, 5]. It overtook both SARS and MERS outbreaks in terms of infectivity and spread [6]. This crisis still poses a threat to public health and socioeconomics globally ever since its appearance in late 2019. SARS-CoV-2 is from a family of viruses that undergo frequent mutations that alter its characteristics like infectiousness and the rate of transmission.

To control the spread of the pandemic, expanded and rapid point of care (POC) testing is of utmost importance. The reverse transcription-quantitative polymerase chain reaction (RT-qPCR) molecular testing is the gold standard according to world health organization (WHO). This bulky and expensive instrument requires special facilities and must be operated by experts, which makes expanded testing challenging in resource-poor regions. Alternative testing methods are needed for simple and fast POC testing to resolve the demand for reagents and diagnostic equipment in such regions. Therefore, fast, affordable, and practical alternative diagnostic tests to detect the novel SARS-CoV-2 are being investigated, one of which meets these criteria is the loop-mediated isothermal amplification (LAMP). LAMP is a molecular test that provides simple detection methods including the colorimetric LAMP, which relies on a color change in the presence of viral nucleic acids and can be detected via the naked eye. Addition of reverse transcriptase (RT-LAMP) enables the detection of RNA pathogens in a one tube reaction. This single-step technique does not require sophisticated equipment, as the reaction takes place isothermally. This chapter summarizes the most recent developments on the colorimetric RT-LAMP assay for the diagnosis of COVID-19.

2. Principle of RT-LAMP

Founded by Notomi [7], RT-LAMP is a molecular diagnostic technique that detects and amplifies nucleic acids (DNA or RNA) using 4–6 primers that identify six regions in a gene, these include forward and backward inner primers and forward and backward outer primers designated as FIP, BIP, F3, and B3, respectively, in addition to forward and backward loop primers (LF and LB) to increase the amplification rate. Unlike RT-qPCR, RT-LAMP is a single-step isothermal reaction that takes place in a heat block or water bath heated to 60–70°C (depending on the enzyme used), thus eliminating the need of a thermal cycler, and the results can be obtained as early as 15 minutes [8]. This technique caught the attention of scientists during COVID-19 pandemic because it provides a cheaper POC testing alternative while maintaining sensitivity and specificity.

RT-LAMP reaction takes place in a single tube containing reverse transcriptase, DNA polymerase, and 4–6 primers. It starts by converting the viral RNA to cDNA via reverse transcriptase enzyme. Then, FIP anneals to its complementary region, initiating DNA synthesis via DNA polymerase with strand displacement activity. Subsequently, F3 anneals outside FIP to its complementary region and starts polymerization and displaces the strand synthesized by FIP. This released strand has a complementary region at 5` end and thus forms a stem-loop structure, which becomes a template strand for BIP. At the other end, BIP binds to its complementary region and similarly starts polymerization, forming a new strand that is displaced by B3. The released BIP strand has complementary sequences at both ends that form a dumbbell structure, which serves as a starting point of LAMP amplification cycle at loop regions by inner and loop primers. This repeated annealing and displacing cycle rapidly releases millions of the targeted DNA structures with different lengths and inverted repeats (Figure 1) [7].

Figure 1.

Main components and primers binding mechanism of loop-mediated isothermal amplification (LAMP) reaction. For the detection of SARS-CoV-2, the reaction tube must be composed of reverse transcriptase, DNA polymerase, primers, and SARS-CoV-2 RNA template, in addition to the reaction building blocks and cofactors like dNTPs and MgSO₄. The amplification starts by binding the forward inner primer (FIP) to its complementary target sequence that results in synthesizing a new strand via DNA polymerase. FIP is then displaced by the outer forward primer F3 annealing and similarly starts polymerization. The released FIP becomes the template strand in which both backward inner and outer primers (BIP and B3) bind and start polymerization and strand displacement as well. Both FIP and BIP form dumbbell structure that become the starting point of loop primers amplification cycle, resulting in millions of DNA target sequences with different lengths and inverted repeats.

Many studies proved the high specificity of RT-LAMP against SARS-CoV-2 [9, 10, 11, 12, 13]. RT-LAMP products can be detected by different methods including the colorimetry, fluorometry, turbidity, and gel electrophoresis, with the colorimetry being the most attractive for its accessibility since it does not require additional instruments for the results’ interpretation. Researchers are developing colorimetric RT-LAMP kits to be commercially used for COVID-19 in vitro diagnosis (IVD) in hospitals, airports, or in-home self-testing. These kits provide a fast and simple diagnosis that can be visualized by a robust color change between positive and negative specimens. For instance, FastProof™ 30 min-TTR SARS-CoV-2 RT-LAMP is a kit that obtained an FDA approval. This kit targets the N gene and gives the results within 30 minutes <https://zenosticbio.com/fastproof-30-min-ttr-sars-cov-2-rt-lamp-kit/>. Several studies validated the performance of this kit. For example, Promlek et al. collected 315 clinical specimens from a hospital during the fourth wave of COVID-19 in Thailand. The assay had a 100% positivity rate in specimens with Ct < 31 in RT-qPCR. However, the accuracy decreased for samples that had higher Ct values. They revealed that the kit was able to detect the infection with high accuracy, particularly in symptomatic patients [14]. Also, the Color SARS-CoV-2 RT-LAMP Diagnostic Assay has been authorized for emergency use by FDA. The FDA recommends using an internal control in these assays to assure a successful reaction [15]. A five-month multicenter prospective observational study evaluated the diagnostic accuracy of the colorimetric RT-LAMP in resource-limited areas in parts of Africa. The study included 1657 symptomatic and asymptomatic individuals and the results were compared with the gold standard RT-qPCR. They obtained 87% overall sensitivity that reaches up to 98% in high viral load samples (Ct < 30) [16]. These results suggest that the colorimetric RT-LAMP is reliable and applicable for POC testing, especially in poor resources areas for offering a cheaper alternative to the gold standard, with a comparable accuracy.

3. Colorimetric RT-LAMP in the detection of SARS-CoV-2

The colorimetric RT-LAMP is one of the simplest detection methods offered in this technique because it can be qualitatively visualized by the naked eye. The color change in the solution post-reaction is due to the accumulation of pyrophosphate ions (PPi) produced during amplification, which causes a pH drop. On average, the pH value drops from 8 to 6 before and after the colorimetric RT-LAMP reaction, respectively [9]. The pH-sensitive dyes that are capable changing color around the reaction pH include phenol red, neutral red, and cresol red. These dyes can be added to the reaction as an amplification indicator. Also, some DNA intercalating dyes such as propidium iodide (PI), SYBR green, SYTO-9, etc., can show a distinguished change in positive samples both colorimetrically and under UV visualization. Additionally, the color change can be quantified using the spectrophotometer via measuring the absorbance at 434 nm and 560 nm wavelengths.

3.1 pH-sensitive dyes

The selection of the pH-sensitive dye is crucial for the best visual detection. The color transition zone of the dye should cover the initial and final reaction pH. In general, when using a minimal reaction buffer (26 μM Tris), the pH should be adjusted to 8.8 with 1 M KOH [17]. After 1 h incubation at 65°C, the estimated pH drops around 6.0–6.5, which is a > 2 pH unit drop due to acidification. The dye stock solution can be prepared in large quantities (50–100 mM) with distilled water and should be diluted to 2.5 mM (25×). However, the dye concentration to be used in the reaction mixture is changeable according to the dye type. In general, a final concentration of 100 μM provides more striking visual detection for most of the dyes. However, cresol red demonstrates more obvious color difference when used at lower concentration (50–100 μM) [18]. Before the reaction, the pH of the reaction solution should be adjusted above the indicating threshold [17]. Therefore, only the dyes that react to the pH of the reaction mixture can be used in the LAMP reaction; these dyes are phenol red, neutral red, bromothymol blue, m-cresol purple, and cresol red. Other than these dyes, a higher pH requiring dye (>pH 9.5) such as naphtholphthalein and thymol blue can be used in visual detection. However, the reaction pH should be adjusted to a higher pH value than usual. In this case, the reaction takes longer time due to the suboptimal pH conditions for polymerase and reverse transcriptase enzymes. The most suitable pH-sensitive dyes with corresponding color changes at low and high pH, and transition pH range for colorimetric RT-LAMP are indicated in Table 1.

Dye	Final conc.	Color change	Low pH color	High pH color	pH range
Phenol Red	100 μM	Red to Yellow	Yellow	Red	6.8–8.2
Cresol Red	50 μM	Red to Yellow	Yellow	Reddish-purple	7.2–8.8
Neutral Red	100 μM	Yellow to Red	Red	Yellow	6.8–8.0
m-Cresol Purple	100 μM	Purple to Yellow	Yellow	Purple	5.2–6.8
Bromothymol Blue	100 μM	Blue to Yellow	Yellow	Blue	6.0–7.6

Table 1.

pH-sensitive dyes for the colorimetric RT-LAMP reactions.

Recently, Amaral et al. [19] suggested a new assay to ease the distinguishing of color by naked eye. A complexometric indicator, namely murexide (MX), obviously forms pink color in positive samples, while remaining yellow in negative ones. The working principle of the dye based on the complexation of MX (2 μL of 5 mM) with Zinc ions (Zn⁺²) (1 μL of 50 mM of ZnCl₂) that are added to the post-reaction (30 min). In the presence of Zn⁺², MX possesses yellow color (negative); whereas it turns to pink (positive) in the absence of Zn⁺². Pyrophosphates (PPi), as the byproduct of LAMP, strongly interacts with Zn⁺² ions and prevents MX-Zn⁺² interaction. This competition leads to a color formation in LAMP-positive samples as pink.

3.2 Intercalating dyes

In addition to pH-sensitive dyes, intercalating dyes such as propidium iodide (PI), ethidium bromide, SYBR green, SYTO-9, Eva green, and GeneFinder can be used to distinguish positive or negative test results. These dyes are added to the reaction mix at the end of the reaction and visualized either by the naked eye or under UV trans-illuminator. For example, for the visual inspection, PI at a concentration of 1 mg/mL can be mixed to the end point RT-LAMP product. Although the use of these dyes facilitates the interpreting of results, they can lead to results misinterpretation owing to the abundant primer concentration of the reaction mixture. In negative or non-template control (NTC) samples, binding of the dye to the primers and non-specific amplicons leads bright transillumination under UV, which makes distinguishing between positives and negatives difficult. Also, the addition of dye at the end of the reaction necessitates opening the reaction tube, which maximizes the carry-over contamination risk. To minimize it, some studies have suggested adding the dye to the tube cap prior to the reaction, this way it can be called a cap-tube assay.

Yu et al. [20] used GeneFinder™ (D039 from Bridgen) to improve the fluorescent signal and sensitivity of the RT-LAMP. The SARS-CoV-2 positive samples are observed as green under blue light, while it was pink in the negative samples. Addition of 2 μL of SYBR Green I (1:10 dilution in TAE buffer) or 0.5 μL of 50× SYBR green into the positive reaction tubes (25 μL) changed the solution color from orange to yellow [21]. In addition, the positive tubes can be observed under UV and/or by loading the product in agarose gel. A ladder-type banding pattern is the typical positive RT-LAMP result.

In addition, dye combinations can improve the visual observation. For instance, adding PI dye to the end-point reaction of neutral red-stained LAMP improves the color difference (data not shown). Moreover, the addition of PI enables the detection of the reaction under UV, which helps to verify colorimetric results in the case of intermediate results. Pre-addition of colorimetric fluorescence indicator (CFI), a color mixture composed of 0.7% (v/v) 10,000× Gelgreen (Biotium, Freemont, 108 CA) in 12 mM Hydroxynapthol blue (Sigma-Aldrich, Oakville, ON) re-suspended in dH₂O, changes the reaction color from orange to yellow in positive samples. The color change can be detected by the naked eye after excitation of gel green in the reaction with a blue LED light. The sensitivity of the assay was 50 copies/reaction, and the reaction time was 30 min by targeting the S gene [22].

Another study by Zhang et al. [23] reported that the addition of guanidium chloride improves the sensitivity for colorimetric and fluorometric SARS-CoV-2 detection about five to ten folds. Moreover, addition of 40–50 mM of guanidine chloride solution (pH ∼ 8 adjusted with KOH) speed up the LAMP reaction time about two-fold, corresponding to ∼40% improvement. Additionally, no nonspecific amplification was detected after the incubation period, which enables better colorimetric and fluorometric LAMP discrimination. The mechanism of action of guanidine hydrochloride in the RT-LAMP reaction has not yet been clarified. According to Zhang et al., [23] guanidine hydrochloride improves the base pairing between primers and target sequences. The addition of guanidine also significantly shortens reaction times for helicase-dependent amplification. The increase is probably not the result of modulation of enzyme activity, as additional enzyme such as reverse transcriptase or Bst 2.0 DNA polymerase had no such effect.

4. Selection of target genes

SARS-CoV-2 genome is comprised of a positive and single-stranded ∼30 kb in length RNA. The subgenomic RNA (sgRNA) encodes conserved structural proteins (spike protein [S], envelope protein [E], membrane protein [M], and nucleocapsid protein [N]), and several accessory proteins. In addition, non-structural protein coding genes such as nsp12, which harbors RNA-dependent RNA polymerase (RdRp) gene responsible for replication, reside in the viral genome. ORF1ab as the largest gene, encodes for two polyproteins PP1ab and PP1a, can also be a target. Dong et al. [24] reported that an N gene-based RT-LAMP assay is more sensitive in detecting SARS-CoV-2 than those based on other genes. Among the target genes, N gene possesses the most abundant expression of subgenomic mRNA during infection [19, 25]. Therefore, it can be the most suitable target for LAMP reaction, especially in the low viral loaded specimens. In parallel, it was reported that RT-PCR assay detected single gene N, but not orf1ab gene upon 9–10 days from the onset of the disease. After 2 weeks, both genes resulted in positive amplification, pointing out the sensitivity of N gene over orf1ab [26, 27]. It should be also noted that because of the high-frequency mutation rate of the SARS-CoV-2 genome, the sequence of the target genes is evolving. A recent study estimated that the nucleotide mutation rate of the SARS-CoV-2 genome is 6.677 × 10⁻⁴ substitution/site/year [28]. For instance, the S gene of the Omicron variant harbors double number of mutations compared to that of the Delta variant, which caused 29 amino acid substitutions and one insertion mutations [29]. Due to this dynamic nature of the genome as well as the viral recombination, the genome is very prone to mutations, which can lead to loss of assay’s sensitivity [30]. This requires at least monthly checking and updating of existing primer sequences.

In addition to selecting a single mRNA target, some studies have reported improved RT-LAMP sensitivity upon using two gene targets simultaneously. According to Zhang and Tanner [31], combining two primer sets that target the same or different genes in the same LAMP reaction led to a higher sensitivity (∼12.5 SARS-CoV-2 RNA copies/reaction) compared to the reactions used one primer set. In addition, some dual primer sets exhibited higher sensitivity at lower temperatures (60°C). However, utilizing the LAMP reaction with triple primer sets instead of dual ones showed no advantage over dual sets.

5. Advantages and drawbacks of the colorimetric RT-LAMP

The colorimetric RT-LAMP technique is gaining popularity in diagnosing a variety of pathogenic diseases, one of which is the most recent COVID-19 pandemic. Due to the large number of infections per day, the global demand for reagents, facilities, and healthcare workers increases, and thus RT-qPCR falls short under these circumstances. The simplicity and accessibility of the colorimetric RT-LAMP enables testing a large number of people in a shorter time without the need of complicated instruments like a thermal cycler. Also, since this test does not require advanced laboratory setting and experienced personnel to be operated, it can be used in rural areas with low resources or POC settings like airports and emergency departments. In addition, the use of four to six primers improves the amplification selectivity for the target sequence, and thus this technique is highly specific for SARS-CoV-2 and not for other viruses, including respiratory and other coronaviruses. Regardless, the results from RT-LAMP must be validated with RT-qPCR when developing new assays to assure high sensitivity [11].

One of the most common limitations in this technique is the formation of unwanted primer secondary structures that lead to misamplification in negative samples, which leads to false positive diagnosis. Therefore, primers must be well designed to avoid forming these structures. Some open-source LAMP primer design platforms have been developed such as Primer Explorer V5 by Eiken (https://primerexplorer.jp/e/), LAMP Primer Design Tool by NEB (https://lamp.neb.com/). However, the classical empirical testing of primers often yields suboptimal results. To acquire the best sensitivity and specificity, several LAMP primer sets should be designed [31]. For instance, Yang et al. [32] tested 35 sets, and Joung et al. [33] designed 29 primer sets that target different genes. Recently, 18 primer sets, mostly from previous publications, were screened to find the most sensitive primer set. Based on the sensitivity, they were classified as sensitive (eight sets), medium (seven sets), and poor (three sets) [31]. In general, only a few sets that possess a satisfactory result are further suggested as a robust method. The others result in lower sensitivity or false-positive results in NTC. To tackle this issue, some algorithms have been developed to find out the robust primer set. Recently, Huang et al. [34] developed an algorithm and suggested to shorten the FIP and BIP primers, which are the longest ones that have a poor annealing efficacy than other primers. The shortening of these primers led to a significant time-to-threshold (TT) improvement in comparison to that of non-truncated ones. According to the suggested protocol, the optimal stem length was in between 12 and 17 bp with a melting temperature (Tm) of >45°C. By means of this strategy, the sensitivity of the RT-LAMP assay has been improved to detect 1.5 copies/μL of SARS-CoV-2 particles in saliva [34].

Together with the abovementioned limitations, the sensitivity of the colorimetric RT-LAMP decreases when testing samples with low viral loads. Some studies reported indeterminate color change—between positive and negative—in lower viral copies samples [9, 10]. The sensitivity also depends on the days from which symptoms of the disease appeared. It was reported that 5 days was the optimum time in which this technique gives the most accurate results, whereas sensitivity reduces if more than 7 days since the onset of the symptoms have passed [10]. To overcome these limitations, studies suggested using internal controls to evaluate primers’ performance and confirm the success of reaction. In addition, it was found that skipping RNA extraction step, especially in saliva specimens, affects the pH of the solution, leading to false diagnosis. Therefore, RNA extraction step is crucial to assure the accuracy of the colorimetric RT-LAMP results [12].

6. Conclusions and future prospect

Since its emergence, RT-qPCR was the only approved gold standard molecular diagnostic technique that detects SARS-CoV-2. However, once studies started to utilize the colorimetric RT-LAMP for a faster and cheaper alternative, they proved the success of this technology in diagnosing COVID-19. Hence, some developed RT-LAMP SARS-CoV-2 detection kits received the approval by the FDA like the AQ-TOP™ COVID-19 Rapid Detection Kit, the Color SARS-CoV-2 RT-LAMP Diagnostic Assay, FastProof™ 30 min-TTR SARS-CoV-2 RT-LAMP, or the Lucira COVID-19 All-In-One Test Kit Labeling for emergency use authorization (EUA). Thanks to its simplicity and accessibility for end-users, this technology can be easily adapted for custom-use. For instance, Davidson and colleagues developed a paper-based colorimetric assay that contains lyophilized reagents for room temperature storage and distribution [18]. Also, some colorimetric RT-LAMP reagents, especially enzymes, can be made in-house to reduce the cost of the assay. To improve the efficiency of the detection, RT-LAMP can be combined with other technologies like CRISPR [35], nanotechnology [36], and many more. Therefore, the colorimetric RT-LAMP method must be further investigated not only for the detection of SARS-CoV-2, but also for broader applications such as human, plant, or animal-hosted viral pathogens.

Acknowledgments

BioRender < https://biorender.com/> is used to generate Figure 1.

Conflict of interest

The authors declare no conflict of interest.