Considerations for colorblind individuals on selecting colorimetric or fluorescent dye assay outcomes

Isothermal loop-mediated amplification (LAMP) is a simple and rapid form of nucleic acid amplification commonly used in field and laboratory settings [1–3]. Because LAMP takes place at a single temperature, the assay can be performed on a heat block as opposed to a thermocycler, and relatively small amounts of DNA can be detected within 1 h [2,4–6]. Multiple methods of LAMP amplicon detection exist through either real-time detection or end-point detection. Real-time detection can be achieved through the addition of fluorescent dyes or through quenched primers or probes, although the equipment required for real-time detection is often more expensive than using a heat block [4]. End-point detection, which requires less expensive equipment, can be accomplished through the addition of intercalating dyes or using colorimetric dyes that respond to pH [7,8]. Most forms of colorimetric dyes can be added to the reaction master mix prior to the reaction, decreasing the chance of contamination of surfaces with amplicon by opening the tubes after the reaction to add a dye.

As LAMP and other isothermal assays become more common for diagnostic purposes, accessibility issues such as for those with colorblindness may create difficulties in the visualization of results using intercalating and colorimetric dyes. Colorblindness is a sex-linked characteristic carried on the X chromosome and appears mostly in males. Females may carry a defective X chromosome, but the presence of a nondefective X chromosome, acting dominantly, results in full-color vision in those individuals [9]. The most common form of color blindness is red–green colorblindness, which includes four forms: protanomaly, protanopia, deuteranomaly and deuteranopia. Another form of colorblindness is tritanopia and tritanomaly, which is the inability to distinguish between blue and yellow [10]. Deuteranomaly is the most common form of colorblindness, affecting approximately 6% of men and 0.4% of women [11]. Colorimetric dyes may be difficult to interpret for individuals with colorblindness and can make developing an accessible assay difficult [10,12]. In addition, colorblindness may pose difficulties in assays without strong amplification. In this work, the authors investigated the effectiveness of various colorimetric dyes in LAMP and the ability to distinguish between positive and negative for these dyes by colorblind individuals.

Materials & methods

New England Biolabs (NEB) LAMP assay kits (MA, USA) were used to ensure a pH change occurred during the reaction. The LAMP amplification target was an Escherichia coli gBlock (ITD, IA, USA) using primers as described by OptiGene (West Sussex, UK). LAMP reactions were set up using NEB's LAMP components kit per the manufacturer's instructions with a final volume of 25 μl. The master mix comprised 2.5 μl of 10x isothermal amplification buffer, 1.5 μl of MgSO₄ (100 mM), 3.5 μl of dNTP mix (10 mM), 1 μl of Bst 2.0 WarmStart^® DNA polymerase (8000 U/ml), 2.5 μl of 10x primer mix (FIP/BIP 16 μM, F3/B3 2 μM, FLoop/BLoop 4 μM). All amplifications were carried out at 65°C for 32 min on a heat block. All amplification reactions were observed under common fluorescent lighting with the exception of reactions containing SYBR™ green, which were observed under both ambient fluorescent and UV lighting.

LAMP reactions were also set up using NEB's colorimetric kit (NEB) per the manufacturer's instructions with a final volume of 25 μl. The master mix comprised 12.5 μl of WarmStart^® Colorimetric LAMP 2x Master Mix, 2.5 μl 10x LAMP primer mix (FIP/BIP 16 μM, F3/B3 2 μM, FLoop/BLoop 4 μM), 9 μl of deionized water (diH₂O) and 2.5 μl of template. Tubes were interpreted directly after removal from the heat block. Positive samples were yellow whereas negative samples were pink (Figure 1).

LAMP reactions were set up as previously described using NEB's LAMP components kit (NEB) with the addition of 1 μl of hydroxynaphthol blue (HNB; #165660-27-5; Chem-Impex International, IL, USA; 2.5 mM in diH₂O), 10.5 μl of diH₂O and 2.5 μl of template. Tubes were interpreted directly after removal from the heat block. Positive samples were blue whereas negative samples were purple (Figure 2).

LAMP reactions were set up as previously described with the addition of 2 μl of malachite green (#2437-28-8; Sigma-Aldrich, MO, USA; 2.74 mM in diH₂O), 3.5 μl of Tris-EDTA buffer (10 mM of Tris-Cl, 18 mM of EDTA), 6 μl of diH₂O and 2.5 μl of template. Once amplification was complete, the reaction tubes were left at ambient temperature for at least 1 h before reading results. Positive samples were light blue whereas negative samples were clear (Figure 3).

LAMP reactions were set up as previously described with the addition of 11.5 μl of diH₂O and 2.5 μl of template. Once amplification was complete, 2 μl of SYBR™ Green I dye (1000× in diH₂O; Invitrogen, MA, USA) was added to the reaction tubes, vortexed and centrifuged. SYBR™ green cannot be added prereaction as it inhibits amplification. Using SYBR™ green, results can be assessed under both ambient and UV light. Under ambient light, positive samples were yellow-green whereas negative samples were orange (Figure 4A). In addition, positive samples fluoresced under UV light whereas negative samples did not (Figure 4B).

Results

The LAMP samples using colorimetric or fluorescent dyes were observed by five adult males with self-reported colorblindness, including one individual with protanopia, one individual with protanomaly, two individuals with deuteranopia and one individual with deuteranomaly. Each individual was asked to identify their colorblindness using the Ishihara Color Test (www.colour-blindness.com/colour-blindness-tests/ishihara-colour-test-plates/). Four people with full-color vision also confirmed the colors in the reaction tubes. Strong color changes were elicited; however, multiple runs were needed to reduce nonspecific amplification in the negative template control (NTC) to ensure the ability to distinguish between positive and negative reactions. The NEB colorimetric kit uses phenol red dye as its indicator. Colorblind individuals were able to distinguish between positive and negative tubes, although the individuals with protanopia, protanomaly and deuteranomaly were not able to accurately identify the yellow-gold color in the positive tubes (Table 1).

HNB dye remained difficult to distinguish for both colorblind and noncolorblind individuals. While those with full-color vision could differentiate between positive and negative tubes, the difference between the blue and violet colors raised concerns over determining positive samples. Four of the five colorblind individuals could not distinguish between the positive and negative tubes for the HNB dye (Table 2).

Most participants experienced little difficulty in distinguishing the colors in malachite green tubes. However, the color change can be subtle, and the individual with protanomaly could not distinguish between positive and negative tubes. In addition, the individual with deuteranomaly misidentified the light teal color as pink. Malachite green tubes were more easily interpreted using a light background, and the NTC in malachite green needs to be clear to distinguish easily (Table 3).

SYBR™ green dye was the easiest to distinguish between positive and negative for colorblind individuals as this dye fluoresces under UV light. Despite this, there was some difficulty in accurately determining the orange and yellow-gold colors in the negative tubes, with one of the individuals with deuteranopia and the individuals with protanomaly and deuteranomaly labeling this color green, although this may be the result of subjective interpretation of colors rather than a misidentification. The individual with protanopia reported the color as red (Table 4). Because SYBR™ green is an intercalating dye, it does not depend on a pH change. Nonspecific amplification in NTC made SYBR™ more difficult to distinguish. Higher primer concentrations also lead to low levels of fluorescence in the NTC.

Discussion

Individuals with red–green colorblindness often have difficulty not only distinguishing between red and green, but colors containing these shades, such as orange and purple. As predicted, colors in the orange and purple spectrums were challenging to interpret for colorblind individuals. While the sample size of self-reported colorblind individuals responding to this study was small, they represented the three most common forms of colorblindness. Although we found that colorblind individuals may be able to interpret the results of orange tubes compared with tubes of other colors, they were often not accurate in discerning the color itself, which warrants consideration when developing an assay and training staff in its performance. Additionally, the use of smartphone applications designed to assist in color recognition might be considered [9]. Fluorescent dyes or dyes with a substantial difference in turbidity between positive and negative could be prioritized over other colorimetric dyes. Overall, of the dyes tested, HNB was the most difficult for our participants to interpret and phenol red and SYBR™ green dyes were less troublesome. However, a caveat in the use of SYBR™ green is that the dye interferes with the amplification so it must be either dried on the side of the cap of the microtube or added after the reaction has completed.

Colorimetric dyes need strong amplification to elicit a pH change. Lower dilutions or suboptimal primer concentrations or designs may fail to elicit a strong enough color change to accurately discern. In addition, colorimetric and intercalating dyes will produce more false positives than a labeled primer or probe. Because LAMP is prone to nonspecific amplification, colors between the positive and negative results are also common, which may result in increased difficulty in distinguishing between positive and negative tubes for colorblind individuals.

Colorimetric dyes can only be used in assays without a strong buffer, otherwise, the pH will not decrease enough for a visual distinction between positive and negative. For example, LAMP kits such as OptiGene cannot be used in conjunction with colorimetric dyes as they are strongly buffered. Some kits being strongly buffered limits the choice of LAMP kits when developing an assay.

Any use of trade, firm or product names is for descriptive purposes only and does not imply endorsement by the US Government.