HRM Analisis

Introduction to High Resolution Melt Analysis

Application GuideHigh Resolution Melt Analysis

Contents Page

Introduction 3

Overview of the Melting Profile Principle 3

The Intercalating Dye Non-saturating dyes 4 Saturating dyes 4 Release-on-demand dyes 4

Instruments and Software 5

Summary of the Workflow Calibration of the instrument/Assay design/PCR/HRM/Data analysis 5

Analysis of Results Amplification plots 6 Raw data melt curve 6 Normalization data 6 Difference graph 6

HRM Assay and Reagent Optimization DNA quality and quantity 7 Primer design 8 Primer concentration and purification 8 Effect of MgCl2 on HRM analysis 9 Optimization of MgCl2 for KAPA HRM FAST 9 Amplicon length 9 Effect of PCR enhancers on HRM analysis 9

Common Problems and Assay Modification Positioning of pre- and post-melt regions within the melt region 10 Non-specific amplification 11 Adjustment of annealing and extension time 12

Specific Applications SNP genotyping 13 Amplification of two or more SNPs on a single amplicon 13 Mutation scanning and screening for heterozygosity 14

Typical Setup for Amplification and HRM with KAPA HRM FAST Amplification and HRM using KAPA HRM FAST 15 Elimination of amplicon contamination prior to assay 15

Troubleshooting 16

Ordering Information 17

Version 1.10


3

Introduction to High Resolution Melt Analysis

Introduction

High Resolution Melting Analysis (HRM or HRMA) is a recently developed technique for fast, high-throughput post-PCR analysis of genetic mutations or variance in nucleic acid sequences. It enables researchers to rapidly detect and categorize genetic mutations (e.g. single nucleotide polymorphisms (SNPs)), identify new genetic variants without sequencing (gene scanning) or determine the genetic variation in a population (e.g. viral diversity) prior to sequencing.

The first step of the HRM protocol is the amplification of the region of interest, using standard PCR techniques, in the presence of a specialized double-stranded DNA (dsDNA) binding dye. This specialized dye is highly fluorescent when bound to dsDNA and poorly fluorescent in the unbound state. This change allows the user to monitor the DNA amplification during PCR (as in quantitative PCR).

After completion of the PCR step, the amplified target is gradually denatured by increasing the temperature in small increments, in order to produce a characteristic melting profile; this is termed melting analysis. The amplified target denatures gradually, releasing the dye, which results in a drop in fluorescence. When set up correctly, HRM is sensitive enough to allow the detection of a single base change between otherwise identical nucleotide sequences.

HRM uses low-cost dyes and requires less optimization than similar systems based on TaqMan® and fluorescence resonance energy transfer (FRET) probes. Compared to these methods HRM is a simpler and more cost-effective way to characterize multiple samples.

This short guide describes the background theory of HRM, in order to assist scientists in the development of HRM protocols. Information is provided on how to optimize assays for maximum sensitivity, how to analyse results obtained when using KAPA HRM FAST, and how to troubleshoot.

Overview of the Melting Profile Principle

Amplification product melting analysis is not a novel concept. Post-PCR melt curve analysis using SYBR® Green I dsDNA intercalating dye to detect primer-dimers or other non-specific products has been performed for over a decade. Today this procedure is called Low Resolution Melting.

A Low Resolution Melt curve is produced by increasing the temperature, typically in 0.5 °C increments, thereby gradually denaturing an amplified DNA target. Since SYBR® Green I is only fluorescent when bound to dsDNA, fluorescence decreases as duplex DNA is denatured. The melting profile depends on the length, GC content, sequence and heterozygosity of the amplified target.

The highest rate of fluorescence decrease is generally at the melting temperature of the DNA sample (Tm). The Tm is defined as the temperature at which 50% of the DNA sample is double-stranded and 50% is single-stranded. The Tm is typically higher for DNA fragments that are longer and/or have a high GC content. The

fluorescence data from low resolution melting curves can easily be used to derive the Tm by plotting the derivative of fluorescence vs. temperature (-dF/dT against T) as shown in Figure 1.

The principle of HRM is the same as a Low Resolution Melt, except that the temperature difference between each fluorescence reading is reduced. During a Low Resolution Melt curve analysis, the temperature increases are typically in 0.5 °C steps, but for HRM this is reduced to 0.008 - 0.2 °C increments. This allows a much more detailed analysis of the melting behavior. HRM sensitivity and reliability has been improved with the use of a variety of new dsDNA intercalating dyes; this is covered in more detail in the following section. The introduction of HRM has revived the use of DNA melting for a wide range of applications, including:

•SNP genotyping•Mutation discovery (gene scanning)•Heterozygosity screening•DNA fingerprinting•Haplotype blocks characterization•DNA methylation analysis•DNA mapping•Species identification•Viral/bacterial population diversity investigation•HLA compatibility typing•Association study (case/control)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

3.2

2.4

2.6

2.8

3

3.2

-dF/

dT

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90Temperature ˚C

Figure 1: Low Resolution Melt profile derivative plot (-dF/dT against T). The steepest slope is easily visualized as a melt peak. In this example the Tm of the amplicon is 77 °C.


4

The Intercalating Dye

There are various types of dsDNA intercalating dyes, which have distinctly different properties. The requirements of the dyes used for HRM are different from dyes typically used for standard quantitative PCR (qPCR) assays. Factors critical in qPCR, such as the signal-to-noise ratio and amplification efficiency, are not essential requirements for HRM. Instead, the dye must provide detailed information on the melting behavior of an amplified target. Ideally the dye should not bind preferentially to pyrimidines or purines, change the Tm of the amplicon, or inhibit DNA amplification. This section briefly describes the three main classes of dsDNA binding dyes that are currently available.

Non-saturating dyes. SYBR® Green I is the most common non-saturating dsDNA intercalating dye. It is generally unsuitable for most HRM applications. At high concentrations, SYBR® Green I stabilizes the duplex DNA and inhibits the DNA polymerase. To allow reliable amplification, low concentrations of SYBR® Green I must therefore be used. At lower concentrations, SYBR® Green I is able to redistribute from the melted regions of single-stranded DNA back to the regions of dsDNA, which results in poor base-difference discrimination, as detailed in Figure 2. To overcome this limitation, saturating dyes have been developed.

Saturating dyes. A new class of dsDNA intercalating dyes that do not inhibit DNA polymerases, or alter the Tm of the product, have recently been developed; these dyes can be added at higher concentrations than non-saturating dyes, ensuring more complete intercalation of the amplicon. The dye is not able to redistribute during melting because the dsDNA is saturated. More precise examination of the melting behavior is therefore possible, as indicated in Figure 2. Dyes such as SYTO9® and LCGreen® can be used at the saturating concentrations required for HRM.

Release-on-demand dyes. The “release-on-demand” class of dyes, which include EvaGreen®, can be added at non-saturating concentrations. This is due to the novel method of fluorescence emission, where the fluorescent signal is quenched when the dye is free in solution. Upon binding to duplex DNA, the quenching factor is released and the dye emits high fluorescent signal. This allows non-saturating concentrations of the dye to be used, ensuring that there is no PCR inhibition, whilst the unique dye chemistry provides highly sensitive HRM analysis.

Figure 2: Non-saturating, saturating and ‘release-on-demand’ dsDNA intercalating dyes. Melting of the duplex as the temperature increases releases the intercalated dyes. At non-saturating concentrations the dye rapidly rebinds to regions that remain double stranded; consequently there is no drop in fluorescence. Saturating and ‘release-on-demand’ dyes do not redistribute from the melted regions of single-stranded DNA back to dsDNA, resulting in a reduction of fluorescence. This difference gives dyes such as EvaGreen® the high sensitivity required for HRM analysis.

Non-saturating dye

No fluorescence change

eg. SYBR

Melting Melting

Saturating dye

Drop in fluorescence

eg. SYT09

Release on demand

Drop in fluorescence

eg. EvaGreen


5

Instruments and software

The power of any HRM analysis depends upon the sensitivity of the instrument and the nature of the PCR reagents used. Certain instruments have been designed specifically for HRM analysis. These can be summarized into two distinct classes:

1. Block based instruments – Samples are placed in a block for cycling and a scanning head or stationary camera is used for detection.

2. Rotary – Samples reside in a single chamber and spin past an optical detector.

There are advantages and disadvantages with each class, and the user should choose the type of instrument that best suits their needs. As a general rule, for high sensitivity and reproducibility a rotary-based instrument should be used, whilst for high-throughput and ease-of-handling, block-type instruments are optimal. The HRM run can vary considerably between instruments; some take considerably longer than others. Every instrument should be calibrated regularly (as recommended by the manufacturer, typically every 6 months), kept in a clean area on a secure and stable benchtop, and checked daily for dust build-up around the optics.

The computer hardware and software must be capable of handling the large quantities of data usually generated during a HRM experiment. Analysis of the data is relatively easy with the appropriate software. HRM specific software uses various algorithms to analyze the data and display the results in easily accessible formats to help discriminate between sequence variants. Some of these algorithms are described in the following sections. Additional software packages, designed for more detailed analysis, are available from the instrument manufacturer.

Summary of the workflow

Pre-assay preparations

Step 1 – Calibration of the instrumentThe correct calibration of the instrument prior to every use is essential, because different dyes have different emission spectra. It is recommended that different manufacturers' products are not used in the same run. For each set of reagents, the calibration of the instrument may require adjustment. The manufacturers’ handbooks describe how to change the filter set to the correct settings prior to PCR and HRM.

Step 2 – Assay designProper assay design is critical for successful HRM analysis. The quality of DNA, correct choice of primers, choice of reagents and optimization of the protocol all play key roles in obtaining the best results possible. The section ‘HRM Assay and Reagent Optimization’ (Page 7) provides detailed instructions on how to design an assay. KAPA HRM FAST PCR Kits are designed for reliable, fast amplification and maximum sensitivity and contain a ready-to-use Master Mix that simplifies the design stage.

Running the assay (Steps 3, 4 and 5 should be completed on the same instrument if possible.)

Step 3 – PCRIt is recommended that real-time PCR is conducted; amplification can then be monitored and any troubleshooting that may subsequently be required is made easier. If an assay has already been optimized, the PCR may be performed on any standard thermocycler (ramp speeds may vary between instruments and the annealing/extension times may have to be adjusted) and the samples then transferred to a real-time thermocycler capable of HRM.

Step 4 – HRMThe HRM step is performed after the final cycle of the PCR. Usually, various settings can be adjusted; these include the start and end temperatures, and the temperature increments and hold times between data acquisitions, typically 0.008 – 0.2 °C steps, with a 2 second hold-time. Some instruments, however, do not allow such a high degree of flexibility and have a single protocol that cannot be adjusted. The section ‘Typical setup for amplification and HRM with KAPA HRM FAST’ (Page 15) describes how to set up a PCR and HRM protocol.

Post-assay analysis

Step 5 – Data AnalysisThe HRM data is analyzed by software specifically designed for the purpose (usually supplied by the thermocycler manufacturer). The section ‘Analysis of Results’ (Page 6) details how to select pre- and post-melt regions for creating normalized and difference plots. The most common problems encountered and how to solve them are summarized in ‘Troubleshooting’ (Page 16).


6

Analysis of results

With the correct software, data analysis is typically straightforward, allowing multiple samples to be analyzed simultaneously. It is important to know what to look for when interrogating results; unforeseen errors made during setup can be identified at this point. Low resolution melt plots are often viewed as a derivative plot (−dF/dT against T); however, HRM raw melting curves must be analyzed differently. This section concerns the analysis of HRM results; a 124 bp product covering a Type IV SNP (rs641805) is used as an example. The most common problems encountered and their solutions are also discussed.

Amplification plots. Reproducibility of amplification is more important for HRM analysis than the efficiency of amplification (as compared to qPCR). For example, HRM analysis can still be performed if the product amplifies later than expected (i.e. due to limitations in primer design), provided that all the samples are consistent, the no template control (NTC) does not amplify and the product is specific.

Raw data melt curve. The data collected during the HRM analysis has a range of initial (pre-melt) fluorescence readings. This variance makes it difficult to properly analyze results, even though different genotype groups may be visible. Figure 3A illustrates how the selection of pre-and post-melt regions is used to align data. Pre- and post-melt regions must be selected for each primer set, by positioning the parallel double-bars as shown. It is important to adjust these bars so that the melt region is not selected. If the pre- or post-melt regions are not clearly defined, it is possible to repeat the HRM run only (without repeating the amplification step), and adjust the temperature range as required.

Normalization data. If the data is normalized correctly, it will appear as shown in Figure 3B. This is termed ‘Normalization Data’. Here the fluorescence variance seen in Figure 3A has been eliminated, and only the temperature range between the outer bars of the pre- and post-melt regions is shown. The genotypes are now more distinct, but the two homozygote samples (in red and blue) are still difficult to distinguish.

Difference graph. Some HRM software applications allow calculation of the difference plot. This is achieved by subtracting the normalized fluorescence data of a user-defined genotype from that of each of the other samples in the HRM analysis. In Figure 3C, the A/A genotype has been selected as a baseline (any genotype can be selected, but usually one of the homozygotes is used). The position of each sample relative to the baseline is plotted against the temperature. This form of HRM analysis is an aid for visualizing the Normalization Data.

Automatic calling of genotypes of sequence variants can usually be performed with the software; sometimes confidence ratings are given. However, checking the difference to confirm the genotype is recommended.

10

0

20

30

40

50

60

70

80

90

100N

omal

ized

Flu

ores

ence

Temperature ( ˚C)74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79 79.5 80

-14

-16

-12

-10

-8

-6

-4

-2

0

2

4

6

8

Nom

aliz

ed m

inus

(A/A

)

74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79 79.5 80Temperature ( ˚C)

Fluo

rese

nce

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84Temperature ( ˚C)

Pre-melt Melt Post-melt

10

0

20

30

40

50

60

70

80

90

100

Nor

mal

izat

ion

regi

on 1

Nor

mal

izat

ion

regi

on 2

Fluorescence variance

Figure 3: Analysis of HRM data from a type 4 SNP (A/T). Different genotypes are highlighted in different colors. A/A = red, A/T = green, and T/T = blue. A. Raw data showing the pre-melt, melt and post-melt regions. Notice the fluorescence variance and the positioning of the pre- and post-melt identification bars. B. Normalization Data derived from the raw data plots in A. Positioning of the pre-melt, and post-melt bars provides a more detailed view of the melt region. C. Difference Graph derived from the Normalization Data.


7

HRM assay and reagent optimization

Due to the highly sensitive nature of HRM analysis, proper assay design is critical. Particular attention should be given to primer design, PCR reagents and cycling conditions. Small differences in the initial setup can greatly affect the final results. Factors such as genomic DNA (gDNA) quality, primer design, MgCl2 concentration and carryover of inhibitors will all affect the melting behavior. Achieving specific amplification is critical to the success of an assay, since any non-specific amplification will greatly impair the melt analysis. The following sections address each of these concerns.

DNA quality and quantity. One of the major factors that can detract from the quality of results is the carryover of buffer from template DNA samples/preparations. Unsuitable DNA purification techniques can exacerbate this problem. Salt carryover can subtly change the melting behavior of the PCR product, and can result in low sensitivity, poor reproducibility and incorrect genotype calls. Buffer carryover from the template DNA will not only modify the Tm and melting behavior during the HRM, but can also encourage non-specific amplification during the PCR.

For optimum results the following is recommended:

•Ideally, all DNA samples to be analyzed (including genotype controls) should be extracted using the same DNA extraction method or kit.

•Some kits use high concentrations of salt for elution of DNA from columns. All template samples should be eluted with low salt buffers, ideally 10 mM Tris-HCl, pH 8.5.

•The DNA sample should be as concentrated as possible after purification. The template can then be diluted with 10 mM Tris-HCl, pH 8.5 prior to use. This helps to reduce buffer variation between samples.

•Addition of a low volume of template DNA will reduce the carryover salt in the reaction; ideally the lowest volume of template that can be accurately added should be used (1 - 2 μl).

•All samples should have a similar final concentration of DNA in each assay. Table 1 details the recommended concentration of DNA for different types of template. At low template concentrations, there is a greater chance of incorporating a mutation early in the PCR which will affect the melting behavior; high concentrations of DNA will result in high background fluorescence due to increased intercalation of dye into dsDNA.

•Positive control(s) for all genotypes should be included where possible, preferably at the same concentration as their corresponding test samples. Control DNA should also be eluted and/or diluted in the same buffer as the samples.

Type of DNARecommended amount

for HRM analysis/reaction

Genomic DNA (gDNA) 10 ng - 100 pg

Plasmid DNA (1 – 10 kb) 1 ng - 10 fg

Amplicon DNA 10 pg - 1 fg

Table 1

Related product: KAPA Express Extract Extraction of HRM-ready DNA by a quick and simple protocol

KAPA Express Extract is a novel thermostable protease and buffer system formulated for the extraction of PCR-ready DNA in 10 minutes, from a broad range of tissue types. Combining KAPA Express Extract and KAPA HRM FAST allows for the extraction, amplification and HRM in less than 2 hours.

In order to minimze the effect of salt carryover, the use of KAPA Express Extract Buffer at a final concentration of 0.5X is recommended. For optimum performance, 1 μl of the DNA extract is added directly to each KAPA HRM FAST reaction.

DNA extracted from samples such as buccal swabs and hair follicles (which contain a lower concentration of inhibitors) may be used directly, with no additional purification required. For more complex samples, such as formalin-fixed, paraffin-embedded (FFPE) tissues, the carryover of contaminants can affect the background fluorescence, and will vary between samples. In these cases the extracted sample should be diluted 1:10 – 1:100 with 10 mM Tris-HCl pH 8.5 prior to amplification/HRM.

For more details see http://www.kapabiosystems.com


8

HRM assay and reagent optimization (cont.)

Primer design. The design of primers for HRM analysis can be a challenge, due to the fact that small amplicon sizes are required for maximum sensitivity. The short amplicon lengths required for HRM restricts the number of potential primer locations; well-designed primer sets will improve the quality of the final result. It is common practice to use a primer design software program such as Primer3 or Primer3Plus. This software assists in positioning primers such that specific regions are included or excluded, and amplicon size can be specified. Potential primer sets and full-length amplicons can then be tested for specificity using a BLAST search. Factors such as the position of the primer, the need for a GC region on the 3’ end, the presence of any secondary structures in the primers or target, and the predicted specificity of amplification should be carefully considered. Figure 4 shows an example of the input page of the Primer 3 design package, when used to design primers for amplification of a human SNP.

After the primers have been designed, the following factors must be checked:

•The presence of a GC ‘clamp’ on the 3’ end. This is not essential, but usually improves amplification.

•The specificity of the primers and the presence of known sequence variations. A BLAST search with both primer sets and the expected amplification product should be performed. If possible, sequence variations should also be checked.

•The secondary structure of the primers and the expected

amplification product. This should be examined using appropriate software. Amplification of any secondary structure within the product may result in unusual melting profiles. Primers may need to be redesigned to avoid areas of secondary structure.

•Amplicon size. This must be optimal for the specific application, typically between 50 - 300 bp.

Primer concentration and purification. The primer concentration should not greatly affect the overall HRM result. Minor changes are possible, but are likely to be amplicon specific. Primers that have been purified using standard desalting methods can be used for many HRM applications. The primary consideration is the carryover of salt, and the effect this has on HRM analysis. To minimize the impact of batch to batch variability often observed in standard desalted primers, low primer concentrations (0.05 or 0.1 μM final) can be used. Despite the expense, HPLC-purified primers are always recommended to ensure maximum reproducibility and sensitivity when attempting to differentiate products with very similar melting profiles, e.g. Type IV SNPs

At high primer concentrations, the chance of non-specific amplification is increased due to mispriming. The recommended primer concentration when using KAPA HRM FAST is between 0.05 and 0.5 μM. An example of non-specific amplification is provided in section ‘Non-specific amplification’ on Page 11.

Figure 4: Design of primers for amplification of a human SNP using Primer 3 (http://frodo.wi.mit.edu/primer3/). The target SNP is shown in square brackets (e.g. ... TTAA[A]TGTC...) and the product size is set to 50 – 300 (bp). Selection of the mispriming library ‘HUMAN’ helps reduce the chances of non-specific amplification.

Select appropriate library

type to reduce mispriming

Sequence input

Target SNP (with square

brackets)

Product size (bp)


9

HRM assay and reagent optimization (cont.)

Effect of MgCl2 on HRM analysis. The concentration of MgCl2 in PCR is known to greatly affect amplification efficiency. For some targets, adjusting the MgCl2 concentration will result in reduced non-specific amplification, allowing clearer distinction of sequence variations; however, MgCl2 concentration has other, notable effects on HRM. It is important to realize that the optimum MgCl2 concentration for amplification efficiency is not necessarily optimal for HRM sensitivity. The effect of MgCl2 is shown in Figure 5. As the MgCl2 concentration is increased, the HRM difference graphs change distinctively.

Optimization of MgCl2 for KAPA HRM FAST. KAPA HRM FAST Master Mix does not contain MgCl2, and it is recommended that the MgCl2 concentration be optimized for each new assay prior to large scale analysis. This will help minimize non-specific amplification, assist in separating genotypes and result in a greater likelihood of differentiating sequence variations. An initial titration of MgCl2 at 1.5, 2.5 and 3.5 mM, using the 25 mM MgCl2 provided in the kit, is recommended and further optimization may be required. If this is not possible, a final concentration of 2.5 mM is recommended for all assays.

Amplicon length. The optimal length of an amplicon is dependent upon the nature of the assay. The longer the amplicon,

the more difficult it is to obtain clear discrimination between sequence variants. For single base changes or SNPs, and for single base insertions/deletions, it is recommended that amplicons be 50 – 300 bp. The clearest discrimination is usually seen with the shortest amplicons. As the amplicon size increases, a more complex melting pattern can develop, making it more difficult to distinguish between single nucleotide variants. The use of smaller amplicons may result in low fluorescence values, presumably due to reduced dye incorporation. However, this is usually not a problem unless the assay is run alongside a longer amplicon assay.

When screening for unknown sequence differences, longer amplicons (typically 200 – 500 bp) can be used. This is useful in gene scanning or determining the variation within a population (e.g. viral) and reduces the number of primer sets required.

Effect of PCR enhancers on HRM analysis. PCR enhancers are commonly added to end-point PCR in order to help reduce non-specific amplification, reduce the number of required cycles and increase yield. These enhancers function by assisting the melting and annealing of primers and templates, and can affect HRM performance. For GC-rich amplicons, the addition of DMSO appears not to affect the HRM significantly, and is recommended if amplification is difficult. Overall, the addition of PCR enhancers is not recommended as they usually detract from HRM performance.

0-7-6-5-4-3-2-10123456789

1011121314151617181920212223

Nom

aliz

ed m

inus

T/T

70 70.5 71 71.5 72 72.5 73 73.5 74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79 79.5 80Temperature ˚C

Increasing MgCl2 concentration

Figure 5: Effect of increasing MgCl2 on melting behavior. Three Difference Graphs are overlayed to demonstrate the dependence of amplicon melting on MgCl2 concentration. The effect of MgCl2 is most pronounced in the heterozygote samples (blue), a result of the magnesium stabilizing the annealing of mispairs. The MgCl2 concentrations are 1.5 mM, 2.5 mM and 3.5 mM final. G/G = green, G/A = blue, and A/A = red.


10

Common problems and assay modification

When analyzing results, it is important to know what to expect when something goes wrong. Identification of a problem is vital to help the user to re-design the protocol or alter the parameters of analysis to obtain maximum sensitivity and reliability. In this section, a series of experimental results that have been analyzed inaccurately, or are from poorly optimized protocols, is shown.

Positioning of pre- and post-melt regions within the melt region. When the pre- and/or post-melt regions are positioned within the melting region of the melting curve raw data (as shown in Figure 6A), the subsequent Normalization Data and Difference Graphs obtained are of poorer quality (as shown in Figure 6B and 6C). Placing the post-melting parallel bars within the melting region causes the difference between the two homozygotes to become smaller and less distinct. The problem is solved by repositioning the pre- and post-melt normalization regions outside of the melting region. If the temperature range of the HRM protocol starts or ends within the melting range, repeat the HRM protocol (without repeating the amplification step) using an appropriately wider range to ensure the pre-melt or post-melt regions are clearly visible.

Figure 6: Positioning the post-melt region within the melt curve and subsequent analysis. A. Melting curve raw data showing the position of the post-melt identification bars within the melt region. B. The subsequent Normalized Data when the post-melt identification bars are located within the melt region. C. The Difference Graph does not show as clear differentiation when compared to Figure 3C. Genotypes are typically still distinguishable, but with a large number of samples, the likelihood that some will be mis-called is greatly increased. Similar results are obtained when the temperature range of the HRM protocol ends or begins within the amplicon melting range.

10

0

20

30

40

50

60

70

80

90

100N

omal

ized

Flu

ores

ence

Temperature ( ˚C)74 74.5 75 75.5 76 76.5 77 77.5 78

-14

-16

-12

-10

-8

-6

-4

-2

0

2

4

Nom

aliz

ed m

inus

(A/A

)

74 74.5 75 75.5 76 76.5 77 77.5 78Temperature ( ˚C)

Fluo

rese

nce

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84Temperature ( ˚C)

Pre-melt Melt Post-melt

10

0

20

30

40

50

60

70

80

90

100

Nor

mal

izat

ion

regi

on 1

Nor

mal

izat

ion

regi

on 2

Fluorescence variance


11

Common problems and assay modification (cont.)

Non-specific amplification. Non-specific amplification products generated during PCR may lead to modified melting behavior of the target amplicon. This greatly affects the quality of the final results. It is therefore important to recognize the features of a HRM plot that are indicative of non-specific amplification. To demonstrate the effect of non-specific products, Figure 7 shows the Normalization Data and Difference Graphs from a HRM assay of a Type I SNP, using two different annealing temperatures. Although the three genotypes can still be distinguished at the lower annealing/extension temperature, non-specific products reduce the difference between the variants, increasing the possibility of mis-calling among samples. It is important to note that in some cases, non-specific amplification may occur as a consequence of a single nucleotide difference, which would restrict non-specific amplification to templates of a single genotype.

Various techniques can be employed to reduce non-specific amplification, including the following:

•Increasing the annealing/extension temperature.•Reducing the primer concentration.•Optimizing the MgCl2 concentration.•Using a 3-step protocol with a reduced annealing time and

additional extension step at 72 °C (for longer amplicons typically used in gene scanning or diversity studies (e.g. 200 – 500 bp)).

•Re-designing of the primers.

10

20

30

40

50

60

70

80

90

100

Nor

mal

ized

Flu

ores

ence

7877 79 80 81 82 83 84 85 86Temperature ( ˚C)

-4

-2

0

2

4

6

8

10

12

14

16

18

Nor

mal

ized

min

us T

/T

77 78 79 80 81 82 83 84 85 86Temperature ( ˚C)

10

20

30

40

50

60

70

80

90

100

Nor

mal

ized

Flu

ores

ence

7877 79 80 81 82 83 84 85 86Temperature ( ˚C)

-4

-2

0

2

4

6

8

10

12

14

16

18

Nor

mal

ized

min

us T

/T

77 78 79 80 81 82 83 84 85 86Temperature ( ˚C)

Annealing/extension at 58°C

Annealing/extension at 53°C

bp M C/C C/T T/T NTC C/C C/T T/T NTC

300

200

150

100

58°C 53°C

Figure 7: Effect of non-specific amplification on HRM analysis. HRM of a 106 bp product containing Type I SNP rs2238289 (C:T); this SNP is part of the haplotype involved in determining eye color. The target was amplified using a 30 sec annealing/extension at either 58 °C or 53 °C. C/C = red , C/T = green and T/T = blue. A. 58 °C Normalization Data and Difference Graphs (inset). B. 53 °C Normalization Data and Difference Graphs (inset). C. Samples from each annealing/extension temperature and genotype were run on a 4% agarose gel to determine if non-specific products were produced. NTC = No Template Control, M = KAPA Universal Ladder. An additional very faint band at ~200 bp was present when the annealing/extension temperature was at 53 °C.


12

Common problems and assay modification (cont.)

Adjustment of the annealing/extension time. Obtaining full length products during amplification helps ensure consistent melting behavior in the analyzed samples. When the annealing/extension time is too short, incomplete amplicons (formed when the DNA polymerase does not reach the end of the target) can have a significant effect on the final HRM analysis. To demonstrate the effect of the annealing/extension time on the Normalization Data and Difference Graphs, Figure 8 shows the analysis of a 183 bp product covering a Type I SNP, using two different annealing/extension times. The derivative plots are indistinguishable; however, the Normalization Data and Difference Graphs are distinctly different, with the longer annealing/extension resulting in better separation of the three genotypes.

There are various factors that will affect the optimal annealing/extension time for HRM analysis. These include the length, sequence and secondary structure of the target. Longer annealing/extension times ensure that full length amplicons are produced, but can also increase the possibility of mispriming and non-specific amplification.

10

20

40

30

50

60

70

80

90

100

Nor

mal

ized

Flu

ores

ence

74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79Temperature ( ˚C)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

dF/d

T

71 72 73 74 75 76 77 78 79 80 81 82Temperature ( ˚C)

-10

-5

0

5

10

15

20

25

30

Nom

aliz

ed m

inus

(T/T

)

74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79Temperature ( ˚C)

10

20

40

30

50

60

70

80

90

100

Nor

mal

ized

Flu

ores

ence

74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79Temperature ( ˚C)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

dF/d

T

71 72 73 74 75 76 77 78 79 80 81 82Temperature ( ˚C)

-10

-5

0

5

10

15

20

25

30

Nom

aliz

ed m

inus

(T/T

)

74 74.5 75 75.5 76 76.5 77 77.5 78 78.5 79Temperature ( ˚C)

Figure 8: The effect of annealing/extension time on HRM analysis. A. Normalization Data and Difference Graph (inset, C/C = red, C/T = blue and T/T = purple) of Type I SNP rs11755 (C/T), with a 20 sec annealing/extension time. B. Difference Graph of the Normalization Data in A. C. Normalization Data and Difference Graph (inset, C/C = red, C/T = blue and T/T = purple) of Type 1 SNP rs11755 (C/T), with a 40 sec annealing/extension time. D. Difference Graph of the Normalization Data in C. The data obtained shows more consistent melting profiles between samples when the annealing/extension time is increased.

20 sec annealing/extension 40 sec annealing/extension


13

Specific applications

The generation and comparison of amplicon melting profiles is a powerful tool in molecular biology. HRM is especially useful since it requires no additional setup (aside from PCR), and melt profiles are obtained directly after amplification, supplying information not obtained from low resolution melt profiles. It is a closed tube assay that allows genotyping without the risk of cross-contamination. Additionally, the HRM step can be repeated several times if required (although the intercalating dye will gradually degrade) to verify results or to test modified settings. After the HRM is complete, the samples can still undergo sequencing (after cleanup) or can be run on an agarose gel. In this section, the most common applications of HRM are described, with focus on assay development and optimization.

SNP Genotyping. Single nucleotide polymorphism (SNP) is the term used to describe a DNA sequence variation of a single nucleotide between members of a species or paired chromosomes in an individual. These variations in the DNA sequence can affect how individuals in a population develop diseases and respond to pathogens, chemicals, drugs, vaccines, and other agents. Most SNPs operate in conjunction with multiple other SNPs to manifest an effect; this means that large populations of individuals need to be examined if detailed and meaningful results are required. In humans, SNPs make up about 90% of all genetic variation and occur (on average) every 100 to 300 bases along the 3 billion base genome. SNP studies can play a role in a variety of fields of study, including:

•Personalized medicine•Disease susceptibility•Drug sensitivity or resistance•DNA variants of complex haplotypes•Endangered species population genetics•Crop and livestock breeding

Usually, a short segment around the SNP in question is amplified; the amplicon is typically 50 – 300 bp in length. The longer the amplicon, the more difficult it is to distinguish the different genotypes, and the greater the chance of including other, unwanted SNPs that will affect discrimination. Ideally, the SNP is located close to the middle of the amplicon. An example of HRM analysis of a human SNP is shown in Figure 9. The expected shift in Tm for each type of SNP (or any sequence variation) is detailed in Table 2.

Amplification of two or more SNPs on a single amplicon. In some cases, two or more SNPs can be located very close together making it impossible to design a primer to amplify a single SNP. It is however possible to amplify a region containing multiple SNPs and perform HRM analyses. This will result in several melting profiles, yet the different genotypes of both SNPs can be determined if all the necessary genotype controls are included. The addition of a probe may help distinguish the individual SNPs if necessary.

1015

5

20253035404550556065707580859095

100105

Nor

mal

ized

Flu

ores

ence

72 73 74 75 76 77 78 79 80 81 8 2Temperature ( ˚C)

-2

0

-4

2

4

6

8

12

10

14

16

18

20

22

24

Nom

aliz

ed m

inus

(A/A

)

72 73 74 75 76 77 78 79 80 81 8 2Temperature ( ˚C)

SNP Class

Base Change

Typical change in Tm

Occurrence in human genome

I C/T and G/A Large (<0.5 °C) 64%

II C/A and G/T Large (<0.5 °C) 20%

III C/G Small (0.2 - 0.5 °C) 9%

IV A/T Very Small (<0.2 °C) 7%

Figure 9. HRM Analysis of the rs12913832 SNP (G/A) near the OCA2 gene that is functionally linked to blue or brown eye color. A 124 bp product was amplified from 1 ng of pure genomic DNA using KAPA HRM FAST with primers designed using Primer3. The three genotypes are clearly distinguished in this assay. The G/G homozygote melts at the highest temperature (red) whilst the A/A homozygote melts at approx 1 °C lower (green). The heterozygote is usually easiest to distinguish with an altered melt profile (blue).

Table 2


14

Specific applications (cont.)

Mutation scanning and screening for heterozygosity. The identification of sequence variants without prior knowledge of the identity or location of the variation usually involves screening of entire genes and their upstream/downstream transcription-regulating regions. To efficiently screen the whole target area, assay design is very important. In general, the following guidelines should be followed:

•Amplicons are typically longer than those used for SNP detection, ranging from ~250 bp to 500 bp in length. A longer amplicon length ensures complete coverage of the target region with a reasonable number of primer sets. The length of the amplicon is selected depending on which mutation(s) are to be detected (e.g. a Type I SNP is easier to detect than a Type IV SNP, and so a longer product can be amplified).

•Primers should not cover regions with known sequence variations; this can lead to preferential amplification of one sequence variant over another.

•Amplicons should overlap by at least 25-30 bp (to cover the priming regions and ensure that all parts of the target region are amplified).

•If there are known ‘hot spots’ or sites of interest, shorter products (100 – 200 bp) are preferable over these regions.

Figure 10 shows an example of the location of nine different amplicons covering a 3 kb gene, including exons, introns and neighboring elements. The average amplicon size is 333 bp, but shorter products are used in a region where there is a high concentration of variations.

It is possible to detect sequence variations with longer products (e.g. >500 bp); however, the variation in the melting behavior of the product decreases as amplicon length increases. Longer amplicons are more likely to form complex secondary structures, which may result in multiple melt peaks. Such multiple peaks are also indicative of non-specific amplification occurring during the PCR, so products may need to be analyzed by gel electrophoresis to confirm specificity.

0.4

0.2

0

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

3.2

3.4

3.6

dF/d

T

71 72 73 74 75 76 77 78 79 80 81 82 83 84˚C

G C C G C C G G

600 bp 200 bp 150 bp 200 bp 400 bp

Figure 10: Design of amplicons for scanning a 3 kbp gene and surrounding regions for sequence variations. A gene consisting of 5 exons (blue) requires several different primer sets to ensure complete coverage of the target region. Amplicons (orange) are designed to overlap, with shorter products targeting regions of interest, or known mutation ‘hot-spots’ (mutations shown by red triangles).

Extra information: Heterozygote samples

With heterozygote samples, a composite of heterozygote (resulting from mispairing) and homozygote duplexes are created. This results in a mixture of four different melting profiles, all combined into one fluorescent signal output. The melting profile plot of a heterozygote sample (in green) often displays a double peak compared to the single peak of the homozygotes (red and blue). The relatively unstable mispaired heteroduplexes melt at a lower temperature than the correctly paired samples, so heterozygotes are easily identified. Addition of higher concentrations of MgCl2 can help increase the number of mis-paired heterozygote samples. This is especially useful when performing gene scanning applications.


15

Typical setup for amplification and HRM with KAPA HRM FAST PCR Kits

KAPA HRM FAST PCR Kits contain the ‘release-on-demand’ green fluorescent dye EvaGreen® (Biotium), which was selected to allow maximum sensitivity with minimal inhibition. The buffer is specifically formulated to maximize detection of Type IV SNPs (A/T). Furthermore, KAPA HRM FAST uses a specialized DNA polymerase that was engineered for fast amplification, saving time and increasing throughput. KAPA HRM FAST is a highly sensitive and reliable product for performing HRM assays.

Amplification and HRM using KAPA HRM FAST. KAPA HRM FAST Master Mix is designed for high performance detection of DNA sequence variations. The kit contains a novel DNA polymerase, engineered via a process of molecular evolution, for fast and efficient DNA amplification in the presence of high concentrations of intercalating fluorescent dyes. This allows for rapid amplification and maximum discrimination between sequence variants. Typical reaction setup and cycling protocol is shown in Table 3 and 4.

KAPA HRM FAST Master Mix does not contain MgCl2. A separate vial of 25 mM MgCl2 is supplied to enable easy MgCl2 optimization. Pilot experiments involving the titration of MgCl2 at 1.5 mM, 2.5 mM and 3.5 mM are recommended, with further refinement if required.

Amplification of the target DNA is performed using standard PCR techniques. To maximize throughput of samples, amplification may be performed on a different instrument before transfer to an HRM-capable instrument for analysis. KAPA HRM FAST PCR Kits are stable for up to 1 year if stored properly at -20 ºC. Once a reaction has been set up, the EvaGreen® dye in KAPA HRM FAST is stable for ~24 hours, provided the samples are kept at 4 °C and protected from light. Typically, the HRM is performed immediately after the amplification step is complete. A post-amplification melt at 95 °C for 10 – 60 sec will help the annealing of heterozygote samples and ensure all amplicons are properly melted prior to HRM.

Elimination of amplicon contamination prior to assay. In an uncontrolled environment, it is relatively easy to contaminate reactions with amplicons from a previous run. Since HRM is very sensitive, amplicon contamination will result in the mis-calling of sequence variations, and manifest as mixed or heterozygote samples. For this reason, KAPA HRM FAST includes dUTP instead of dTTP. The inlcusion of dUTP facilitates digestion of any amplicon product with Uracil DNA Glycosylase (UDG) prior to PCR, which effectively eliminates amplicon contamination. UDG is not provided in the kit and must be supplied by the end user.

ComponentRecommended

Volume1Final

Concentration

PCR grade water Up to 20 μl –

KAPA HRM FAST Master Mix (2X)

10 μl 1X

MgCl2 (25 mM)2 2 μl 2.5 mM

Forward Primer3 (10 μM)

0.4 μl 0.2 μM

Reverse Primer3 (10 μM)

0.4 μl 0.2 μM

Template DNA 1 μl100 pg - 20 ng

(gDNA)

Step Temperature Duration Cycle

Enzyme activation

95 °C 3 min Hold

Denaturation 95 °C 5 sec

35 - 45Annealing/Extension

60 °C20 - 30

sec1

Denaturation 95 °C10 - 60

secHold

HRM2 60 - 90 °C - -

Table 3

Table 4

1The annealing/extension time will vary depending on the ramp speed of the instrument. For instruments with a slow ramp speed, a shorter annealing/extension times can be used. For longer amplicons (>150 bp), the addition of an extra 10 sec/100 bp may be required.

2HRM settings are instrument-dependent.

1Reaction volumes may be scaled down from 20 μl to 10 μl if low volume tubes/plates are supported by the instrument.

2Use 2.5 mM final concentration of MgCl2 as a starting point. A final concentration of 1.5 mM - 4.0 mM is recommended for most assays. Refer to page 9 for more information on optimizing MgCl2 for HRM.

3Use 0.2 µM of each primer as a starting point. A final concentration of 0.1 µM - 0.5 µM of each primer is recommended for most assays. Refer to page 8 for more information on optimizing primer concentrations for HRM.


16

Troubleshooting

The problems encountered when performing a HRM assay are usually easily identified. The table below details the most common problems encountered in the PCR and HRM steps, and summarizes previously discussed abnormal data and additional aberrant results. Other resources are available on the Kapa Biosystems website (www.kapabiosystems.com).

Symptom Possible Cause Solution

PC

R

Amplification of the majority of samples have a Ct > 30, result-ing in the reaction not reaching plateau.

Low template concentration •Increase concentration of template DNA.

Poor priming efficiency•Reduce annealing temperature.•Increase annealing time.

MgCl2 concentration is not optimal •Optimize the MgCl2 concentration (e.g. attempt 1.5 mM, 2.5 mM, 3.5 mM MgCl2).

DNA is damaged •Re-purify the DNA.

Ct of samples are inconsistent.

Varying DNA concentrations •Re-check the DNA concentrations.

DNA is damaged •Re-purify the DNA.

Non-specific amplification

Mispriming•Increase the annealing temperature during amplification/extension.•Check for mispriming using a software package and redesign the primers if necessary.•Reduce primer concentration to minimize primer-dimers and non-specific priming.

MgCl2 concentration is not optimal for primer set •Optimize the MgCl2 concentration (e.g. attempt 1.5 mM, 2.5 mM, 3.5 mM MgCl2) for amplification.

Multiple melting peaks appear for one product.

Two products of the same length•Check the products on an agarose gel.•Heterozygosity can result in two peaks.

A HRM melt uses more readings than a standard melt that can give the impression of 2 products on a melting peak

•Check the products on an agarose gel.•Adjust the filter settings of the melt peak analysis to minimize noise.•Increase the annealing/extension temperature to help minimize non-specific amplification.

HR

M

Replicate samples show a wide variance during HRM analysis (i.e. results do not allow accurate geno-typing, particularly be-tween homozygotes).

Variations in reaction mixture (e.g. salt)

•Check the purity of the template solution.•Dilute all template samples in the same buffer (10 mM Tris-HCl, ph 8.5) to equalize salt content.•Re-purify the DNA using a new method.•Ensure all reagents and Master Mixes are vortexed well prior to reaction setup.

Concentrations of template are very different •Ensure all samples are the same DNA concentration for maximum resolution.

Rare or new generic variants may generate results outside of the expected ranges

•Sequence the unusual results.•Redesign primers to produce a shorter amplicon, hopefully avoiding the genetic variation.•Mixing the amplicons (9.5 μl from each of two amplicons) can help identify potential new variants.

Secondary structure in the amplicon•Use a Melting Profile software system to ensure no secondary structure are present (eg. http://mfold.rna.

albany.edu/?q=DINAMelt).

Only one homozygote is detected during HRM.

Difficult to detect mutation (typically Type IV, A/T) •May need to sequence to confirm, or use a probe or snapback primers.

Only one homozygote is present•Mix 9.5 μl of each of the homozygote controls and run a HRM analysis (include an initial melt at 95 °C for

3 min); the melting profile should change to the heterozygote, which is usually easier to detect.

Different genotypes cannot be detected during HRM.

MgCl2 concentration not optimal•Perform amplification and HRM at 1.5, 2.5, 3.5 mM MgCl2 to find the optimal concentration for HRM

analysis.

Non-specific amplification

•Check the melting curve to see if multiple peaks are present (adjust filter settings if necessary).•Run product on a 2 - 3% agarose gel to confirm if multiple bands are present.•Reduce primer concentration to minimize primer-dimers and non-specific priming.•Increase the annealing/extension temperature.•Redesign primers.

Template contains HRM inhibitors (eg. salt) •Re-purify DNA template.

Difficult to detect mutation (typically Type IV, AT)•Shorter amplicons are better for detecting Type III and Type IV mutations.•Redesign the primers to produce a smaller amplicon.•Use a probe or snap-back primers.

Unusual appearance of melt curves.

Multiple melt regions or secondary structure

•Presence of mutation(s) can result in more complex melting curves due to the development of secondary structures. This can still provide reliable HRM data.

•Long amplicons are more likely to develop secondary structures.•Redesign primers to amplify a shorter fragment.

Non-specific amplification•Increase the melting/annealing temperature during amplification.•Optimize the MgCl2 concentration (e.g. 1.5 mM, 2.5 mM, 3.5 mM MgCl2)


17

Ordering information

Please note that KAPA HRM FAST PCR Kits are not recommended for quantitative PCR (qPCR). KAPA SYBR® FAST qPCR Kits and KAPA PROBE FAST qPCR Kits are the recommended kits for qPCR. KAPA Express Extract Kits can be used for fast DNA extraction prior to amplification and HRM analysis using KAPA HRM FAST PCR Kits.

Description Code Kit contents

KAPA HRM FAST PCR Kit



KAPA Express Extract




KK4201 100 reactions

KK4202 500 reactions KK4203 1000 reactions KK7100 50 reactions




www.kapabiosystems.com

Boston, Massachusetts, United States600 West Cummings Park, Suite 2250

Woburn, MA 01801 U.S.A.Tel: +1 781 497 2933 Fax: +1 781 497 2934

Cape Town, South Africa2nd Floor, Old Warehouse Building, Black River Park,Fir Road, Observatory 7925, Cape Town, South Africa

Tel: +27 21 448 8200 Fax: +27 21 448 6503

Email: [email protected]

For more information please contact [email protected] or your local representative.

Purchase of this product includes an immunity from suit under patents specified in the product insert to use only the amount purchased for the purchaser’s own internal research. No other patent rights are conveyed expressly, by implication, or by estoppel. Further information on purchasing licenses may be obtained by contacting the Director of Licensing, Applied Biosystems, 850 Lincoln Centre Drive, Foster City, California 94404, USA.

EvaGreen® is a trademark of Biotium, Inc. Kapa Biosystems is licensed by Biotium, Inc to sell reagents containing EvaGreen® dye for use in HRM, for research purposes only.

The purchase of this product includes a limited, non-transferable license under U.S. Patent No. 5,871,908, owned by Evotec Biosystems GmbH and licensed to Roche Diagnostics GmbH, to use only the enclosed amount of product according to the specified protocols. No right is conveyed, expressly, by implication, or by estoppel, to use any instrument or system under any claim of U.S. Patent No. 5,871,908, other than for the amount of product contained herein.

Licensed under U.S. Patent Nos. 5,338,671 and 5,587,287 and corresponding patents in other countries.

Documents

HRM Analisis