In this article we will discuss about the basics of mutation screening by Denaturing High-Performance Liquid Chromatography (DHPLC).
Introduction to Basics of Mutation by DHPLC:
Studies of sequence variation are an integral part of genome research, because of differences in the genomic sequences between individuals. Variations in mammalian genomic sequences occur at an estimated frequency of at least one base per thousand bases. In the case of the human genome, this totals more than three million variances.
These variances are typically present in the form of single nucleotide polymorphisms (SNP). The relationship between the variability in the sequence of specific genes among individuals to their disease status and/or susceptibility to disease is a primary key to understanding gene function.
Variations, in addition to being critical for identification of disease genes, are also essential in understanding genetic differences in individual responses to the environment, to disease, and therapies.
The challenge of understanding even a fraction of the variations in the genomic sequence, determining which of the polymorphisms are significant and which are inconsequential, is overwhelming. However, the potential value of such knowledge is great in that it will not only lead to better diagnostic and prognostic assays, but also to improved treatment for disease.
A highly sensitive, reproducible, and robust technology for detection of sequence variations is pivotal to exploiting the relationship between sequence variability and function. A recently developed technology for the detection of sequence variance, based on denaturing high-performance liquid chromatography (DHPLC), for the analysis of DNA fragments or PCR products meets the above requirements.
The technology, which is based on the detection of heteroduplexes, allows for automated, high-accuracy identification of SNP as well as small deletions or insertions. Heteroduplex profiles are easily distinguished from homoduplex peaks and thus provide a reliable means for mutation scanning and discovery.
DHPLC has been shown to clearly resolve mutations in various genes with detection rates ranging from 92.5% to 100%. It is reported in the literature that the sensitivity of detection by DHPLC is higher than for alternative gel based analysis techniques, i.e., mutations not detected by gel- based techniques could be detected by DHPLC.
Here, principles underlying mutation detection by temperature modulated heteroduplex analysis (TMHA) using DHPLC technology on the WAVE Nucleic Acid Fragment Analysis Systems from Transgenomic are explained and illustrated. Recent improvements in the WAVE System instrumentation and updates regarding the understanding of the technology are discussed.
Furthermore, the importance of using WAVE Systems compatible samples and elution buffers and of using high-fidelity DNA polymerases for amplification of DNA fragments intended for analysis by DHPLC on the WAVE System are reviewed in the context of success, accuracy, and reliability of the WAVE System technology.
i. WAVE Nucleic Acid Fragment Analysis Systems:
All HPLC analyses shown here were performed on WAVE Nucleic Acid Fragment Analysis Systems equipped with DNASep Cartridges.Three major WAVE System configurations are currently available – 2100A, 3500A, and 3500 HT. In addition, the WAVE-MD System is now available.
This system is based on Transgenomic’s proven, highly sensitive DHPLC/TMHA technology with the DNASep Cartridge and is dedicated for mutation analysis. The WAVE-MD System employs fixed gradient conditions optimized for DNA fragments ranging in size from 100 – 600 base pairs.
Other features include minimal set-up time, continuous and unattended operations, user- friendly WAVE-MD software, and cost effective consumables solutions with a targeted throughput of circa 100 samples per day. Key features of the individual system configurations are listed in Table 1-1.
Chromatography on the WAVE Systems is performed with a dual buffer system, where buffers A is an aqueous solution of 0.1M triethylammonium acetate (TEAA) and buffer B a 0.1 M TEAA solution containing 25% acetonitrile (ACN). CAN,TEAA, and water used for the preparation of buffers were of HPLC grade.
Volumetric flasks were used for the preparation of buffers due to the sensitivity of retention times to even slight variations in ACN concentrations. Alternatively, pre-made 1x buffers A and B (Transgenomic) can be used. Eluent conditions for analysis can be determined using either WAVEMAKER or WAVE Navigator software. Eluent flow rates for all analyses were 0.9 mL/min in the Standard Analysis Mode and 1.5mL/min in the Rapid Analysis Mode.
Chromatograms are generally recorded at a wavelength of 260nm or, in the case of fluorescently labeled DNA, at a wavelength that is characteristic for the fluorescent label used. All analyses under non-denaturing conditions are run at 50°C. Mutation detection under partially denaturing conditions is performed at or near the software predicted temperature indicated for each chromatogram.
ii. DNASep Cartridge Technology:
The DNASep Cartridge is the heart of the WAVE Nucleic Fragment Analysis System. The cartridge contains a robust, nonporous, alkylated poly(styrene- divinylbenzene) separation matrix. Depending on the analysis mode, normal or rapid, a DNASep Cartridge with dimensions of 4.6 x 50mm or 6.5 x 37 mm is used.
The diameter of the DNASep Cartridge for the WAVE System 3500HT in the Rapid Analysis Mode is increased in order to accommodate higher flow rates. The WAVE-MD System utilizes a 4 x 10mm separation cartridge, which is optimized for mutation detection under the standardized conditions of this system.
Under the analysis conditions employed for mutation detection with the WAVE System, the separation matrix exhibits a higher affinity for double-stranded DNA (dsDNA) than for single-stranded DNA (ssDNA). Mutation detection is performed at or slightly below the melting temperature of the homoduplex.
At this temperature heteroduplexes, if present, are partially melted at the mismatch site and therefore exhibit partial ssDNA character. This results in a decreased retention time of partially denatured heteroduplexes as compared to the corresponding homoduplexes.
WAVEMAKER and WAVE Navigator software packages control the operation of the HPLC hardware. Melt profiles of DNA fragments being analyzed for the presence of mutations, analysis temperatures, and HPLC gradient conditions are predicted by WAVEMAKER or WAVE Navigator software.
The analysis temperatures for mutation detection are derived from software predicted melt profiles of percent helical fraction versus base position. Both software packages employ a modified Fixman-Friere algorithm to predict the sequence-dependent melt behavior of DNA fragments.
A temperature at which the target fragment, the PCR product being scanned for the presence of mutations, is 70-85% helical is regarded as the optimal temperature for mutation detection. Both software packages control the acquisition and handling of data. Data analysis features vary depending on the software package used. WAVE Navigator is the latest software product for the operation of the WAVE System and analysis of chromatographic results.
It includes enhanced features over the WAVEMAKER software such as a relational database management system, automated mutation calling, more robust and expansive data storage capabilities, across project data analysis, and remote operation capabilities.
iv. Sample Preparation:
DNA fragments to be analyzed for the presence of mutations are most commonly PCR products, but could also be restriction fragments. Primers flanking the target region for mutation detection are generally designed using conventional primer design software packages such as OLIGO. or Primer 3.
Transgenomic highly recommends that each amplicon sequence be evaluated for its suitability to detect mutations by heteroduplex analysis on the WAVE System using either WAVEMAKER or WAVE Navigator software.
Additionally, a tool for amplicon design is offered for subscribers of an instrument maintenance plan. PCR products amplified from heterozygous templates generate wild type and mutant allele products in equimolar amounts.
PCR products from homozygous templates require post-PCR addition of wild type allele PCR product in order for mutations to be detected. Prior to mutation analysis PCR products need to be denatured and slowly reannealed to promote the formation of heteroduplexes. The latter is generally achieved by heating to 95°C, followed by slow cooling of the sample to 25°C at a rate of minus 0.1°C per minute.
PCR products and restriction digests do not need to be purified prior to mutation analysis on the WAVE System. In fact, purification of DNA fragments prior to analysis on the WAVE System is discouraged, because of a possible introduction of compounds from the purification kit or procedure that may be harmful to the performance and durability of separation cartridge.
Results and Discussion of Mutation Screening by DHPLC:
i. Principles of Mutation Detection by TMHA:
The screening and scanning for sequence variation/mutations by DHPLC involves the analysis of homoduplexes and heteroduplexes that are formed between DNA fragments (PCR products or restriction fragments) derived from wild type and mutant sequences (Figure 1-1). In brief, a primer pair is designed or restriction sites are chosen to flank the DNA sequence of interest. DNA fragments ranging in size from ca. 150 to 1,000bp or longer may be suitable for analysis.
The decisive factors in judging whether a DNA fragment is suitable for mutation detection are its length, sequence, and melt characteristics. PCR in endonuclease digestion of heterozygous templates generates wild type and mutant allele products in equimolar amounts. DNA fragments generated from homozygous mutant template require the addition of equimolar amounts of wild type product to permit formation of heteroduplexes.
Wild type product is generally added when screening for unknown mutations and for the analysis of samples when the occurrence of homozygous mutants is likely or possible. This approach, in the case where the DNA fragment is generated by PCR, is termed mutation detection after post-PCR mixing.
A wild type/mutant mixture is denatured at 95°C. This is followed by slow renaturation, which leads to the formation of wild type and mutant homoduplexes as well as two distinct wild type/mutant heteroduplexes. The latter are characterized by a mismatch at the site where the sequence variance occurs.
Mutation analysis by DHPLC on the WAVE System, which employs the technique of Temperature Modulated Heteroduplex Analysis (TMHA), is based on two key features. First, heteroduplexes always melt at the mismatch site generating a single-stranded, partially denatured region at a temperature below the melting temperature of the corresponding wild type and mutant homoduplexes.
Second, ssDNA is retained less strongly on the DNASep Cartridge than dsDNA. Thus, at a temperature at which wild type and mutant homoduplexes are still partially denatured, resulting in up to two distinct peaks, the wild type/mutant heteroduplexes are already denatured to a greater extent and are less well retained on the column. The more denatured heteroduplexes yield up to two peaks of equal intensity preceding the homoduplex peak(s).
The DNASep Cartridge contains a nonporous, C18-alkylated, and therefore hydrophobic poly(styrene-divinylbenzene) separation matrix. This precludes direct interactions between the hydrophilic nucleic acids analyzed and the hydrophobic separation matrix. Interactions between nucleic acids and the separation matrix are mediated by triethylammonium acetate (TEAA), which is contained in both buffers A and B at a concentration of 100 mM.
Upon sample injection, the cationic triethylammonium ions interact electrostatically, e.g., bind non-covalently, to the anionic nucleic acid phosphate backbone. The alkyl groups of the triethylammonium cations interact with the hydrophobic surface of the alkylated DNASep matrix, mediating the interaction between nucleic acids and the separation matrix.
Eluent buffer A is a purely aqueous solution that is 100 mM in TEAA. Eluent buffer B is 100 mM in TEAA and contains 25 % (v/v) of the organic solvent acetonitrile (ACN). The DNASep Cartridge is charged with a nucleic acids sample at a software predicted ratio of buffers A and B, which ensures binding of the DNA fragments to the column matrix.
Elution of the nucleic acid fragments analyzed, in a manner that is dependent on their lengths as well as their state of naturation/denaturation, is then achieved by a systematic, software-predicted increase in elution buffer B relative to A. Under gradient conditions used for mutation detection nucleotides and primers used in PCR exhibit virtually no affinity for the DNASep matrix.
Consequently, these nucleic acids elute early on in the elution gradient, evidenced by the presence of a large peak in the chromatograin preceding the analysis portion of the gradient (Figure 1-2, peak at 0.5-1.0 min). Polymerases and endonucleases, or proteins at low concentrations in general, are not retained on the separation matrix under the analysis conditions used for mutation detection.
Proteins contained in reaction mixtures are therefore co-eluted with nucleotides and primers at the beginning of the elution gradient. Notable exceptions are extremely hydrophobic proteins, e.g., bovine serum albumin (BSA), at high concentrations, which potentially, accumulate on the column matrix and interfere with column performance, retention times as well as resolution. The elution of the nucleic acids is monitored on line by either UV or, in the case of fluorescently labeled nucleic acids, by fluorescence detection.
ii. Identifying the Presence of a Genetic Variation – Mutation Calling:
Single nucleotide changes as well as short insertions and deletions can be detected with high accuracy and sensitivity by TMHA using either one of the WAVE Nucleic Acid Fragment Analysis Systems or the WAVE-MD System.
Sequence variations are identified based on a comparison between the chromatograms of wild type amplicons, e.g., devoid of sequence variations, and chromatograms of amplicons that carry mutations, which have formed heteroduplexes with wild type amplicons, recorded under identical chromatographic conditions.
A characteristic example of mutation calling is presented in Figure 1-2, A – C. The wild type chromatogram exhibits a typical single-peak pattern, because the amplified wild type sample contains a single homoduplex amplicon. With increasing analysis temperature the PCR product progressively melts, resulting in a steady increase in its single-stranded character and causing a steady reduction in retention time.
The latter is the result of decreased affinity of ssDNA to the column matrix as compared to dsDNA. The single-peak chromatogram of a wild type amplicon, recorded at or near the software-predicted melting temperature of the wild type homoduplex, serves as a reference for mutation calling.
The presence of sequence variations/mutations in an amplified region leads to the formation of four distinct duplexes (Figure 1-1) during the denaturation and reannealing step preceding TMHA. In the case of the wild type/mutant 1 chromatogram (Figure 1-2B), where the mutant is defined as having a C→T change in position 178 of the amplified region, the presence of a sequence variation becomes apparent by the appearance of additional peaks in the chromatograms.
In contrast to the chromatograms for the wild type amplicon, where only one homoduplex is present in the sample (Fig. 1-2A), the 1:1 wild type/mutant 1 mixture contains four distinct duplexes, two homoduplexes that correspond to one wild type and one mutant homoduplex and two heteroduplexes, which are formed by the (+/-) and the (-/+) strands of the wild type and mutant allele PCR products.
Homoduplexes have higher melt temperatures than the corresponding mismatch containing heteroduplexes. Thus, the retention times of peaks representing homoduplexes is always longer than retention times of their heteroduplex counterparts under identical gradient conditions and at analysis temperatures at or near the melt temperature of the homoduplex.
In the wild type/mutant 1 chromatograms presented here, the two heteroduplex species are not resolved from each other at any of the temperatures tested. However, they do elute earlier than the wild type and mutant 1 homo- duplexes.
In this particular case, the single nucleotide change between the wild type and mutant 1 homoduplex is sufficient to result in a difference of melting temperature between the two homoduplexes. Consequently, wild type and mutant 1 homoduplex peaks exhibit differences in retention time. These differences are most obvious at analysis temperatures of 64°C and 65°C.
In the case of the wild type/mutant 2 chromatogram (Figure 1-2C), where the mutant is defined as having an A→G change in position 139 and a C→T change in position 178 of the amplified region, again, the presence of a sequence variation becomes apparent by the appearance of additional peaks in the chromatograms compared to those of the wild type.
Here, heteroduplexes, which are identified based on their shortened retention times, are well resolved from the later eluting homoduplexes. The presence of a sequence variation between wild type and mutant sequence is apparent at all analysis temperatures shown, e.g., the chromatograms obtained for the wild type/mutant 2 sample differ from those of the wild type chromatograms at all temperature recorded.
The software predicted analysis temperature for TMHA on the WAVE System for the amplified sequence investigated here is 63°C. It is noteworthy that the sequence variations were in fact detected not only at the predicted temperature, but also over a temperature interval of several degrees flanking the predicted temperature. While this is not always the case, it is a very common observation.
iii. Mutation Detection in the Rapid Analysis Mode:
With the WAVE Nucleic Acid Fragment Analysis System Models 2100A and 3500A approximately 200 samples can be analyzed per day for the presence of mutations. This corresponds to an average run time per sample, from one injection to the next, of 7.2 minutes. The number of samples analyzed can be doubled using the WAVE System Model 3500 HT.
The increased sample throughput is made possible, in part, by such system modifications as:
(1) Change of the DNASep Cartridge geometry, which permits an increase in flow rate from 0.9 mL/min to 1.5 mL/min,
(2) Use of the smaller eluent mixer, 0.250 mL as compared of 2.00 mL, which supports faster and more precise changes in eluent composition, and
(3) The use of an accelerator, which allows for an accelerated cartridge-cleaning step as well as reduced equilibration times between injections (see Table 1-1). Two examples of mutation detection with the WAVE System 3500 HT in the Rapid Analysis Mode, using Transgenomic’s mutation standards for 56°C and 63°C, are shown in Figure 1-3.
iv. Mutation Detection Using Fluorescently Labeled DNA Fragments:
The use of fluorescent labels is compatible with the technology of mutation detection by DHPLC on the WAVE System and generally increases sensitivity. The affinity of the column matrix for the labeled DNA fragment is increased due to the presence of hydrophobic fluorescent dye molecules.
Increasing the percentage of acetonitrile in the elution buffer can compensate for this effect. Resolution, which is defined as the ability to resolve heteroduplexes from homoduplexes, the ultimate prerequisite for the identification of sequence variance, is not affected.
v. Sample and Eluent Compatibility with the DNASep Cartridge:
It is crucial for consistent, high-quality mutation analyses on the WAVE System that samples and buffer preparations are devoid of contaminants, components, and additives that diminish the performance of the DNASep Cartridge separation matrix. Specific performance factors are resolution and retention of nucleic acid fragments. The separation matrix is C18-alkylated and therefore hydrophobic.
The interaction between the separation matrix and the nucleic acid fragments analyzed is mediated by TEAA. Thus, any compound present at sufficiently high concentrations in the sample or eluent buffers to interfere with the interaction between the separation matrix and the nucleic acid, has the potential to diminish the performance of the DNASep Cartridge.
The damage done to the DNASep Cartridge by some components is reversible, while others cause permanent damage. In any event, compounds known to diminish column performance should be avoided. A list of the classes of contaminants and additives to buffers and samples that are known to diminish column performance is given in Table 1-2.
Small hydrophobic compounds, which are generally introduced with the sample, interact with the hydrophobic column matrix directly and have a negative effect on not only resolution, but also on retention times.
Specific examples of compounds to be avoided, or to be used at minimal concentrations include detergents like Triton X-100, NP40, SDS, or Tween20. Performance of the DNASep Cartridge can usually be fully recovered after contamination with detergents by extensive washing procedures.
Aromatic and aliphatic compounds, which are hydrophobic, can dramatically interfere with column performance. Phenol, which can be carried over after phenol/chloroform extraction of template DNA, does not only interfere with PCR, but also negatively affects DNASep cartridge performance. Mineral oil dramatically and irreversibly alters the performance of the DNASep Cartridge. It is not to be used as an overlay in PCR reactions intended for analysis on the WAVE System.
Certain, mainly polyvalent metal ions, but not Mg2+, will interfere with column performance, e.g., causing severe loss of resolution and variable retention times, because of their ability to bind to the negative phosphate backbone of the nucleic acids analyzed.
Sources of polyvalent cation contamination can be poor quality of the water used for buffer preparation, glassware, or prolonged contact of the eluent buffers with metal parts. The latter can lead to iron contamination, which has a particularly severe effect on column performance.
Macromolecules and particulate material can physically interfere with column performance by obstructing eluent flow and preventing interactions between the column matrix and the nucleic acid fragments analyzed.
Macromolecules, such as large, particularly denatured proteins and carbohydrates, can be carried over with the template or may have been added as enhancers to the PCR, e.g., polyethylene glycol (PEG) or heparin. It is essential to remove macromolecules and particulate material from samples prior to analysis on the WAVE System or to avoid their introduction in the first place.
Several chemicals commonly used as PCR additives and enhancers, such as formamide, glycerol, DMSO, and betaine, should be avoided or at least used with caution, because of their potential to interfere with column performance and/or chemically modify the nucleic acids analyzed. In the case that the use of the aforementioned additives is considered necessary, it is recommended to closely monitor column performance by frequently running mutation standards.
The WAVE System with its DNASep Cartridge provides a robust and reliable technology for the detection of mutations. Consideration of the aforementioned factors affecting column performance will not only ensure consistent, high- quality results, but also long cartridge life.
vi. Importance of Polymerase Fidelity for Mutation Detection:
DNA polymerases inherently incorporate errors randomly during amplification. The extent of error incorporation depends in part on a number of factors, including e.g., dNTP and Mg2+ concentration. However, the primary factor determining the error incorporation rate is the polymerase itself. Polymerases that have 3′ to 5′ exonuclease activity generally referred to as proofreading activity, incorporate the least number of errors and exhibit highest replication fidelity.
It is apparent that errors incorporated at random by the polymerase during amplification have the potential to interfere with the identification of mutations present in the amplified template. Therefore, the use of a high fidelity polymerase such as Transgenomic’s Optimase Polymerase is recommended for the amplification of PCR products to be used for mutation detection.
Representative data on the outcome of different error incorporation rates by several commonly used DNA polymerases and their influence on the ability to reliably call mutations is shown in Figure 1-4. A 500-bp PCR product was amplified from phage genomic DNA using several different polymerases.
PCR products were first analyzed under non-denaturing conditions on the WAVE System at 50°C in order to assess PCR product quality and yield. Results are shown in Figure 1-4, Panel A. As expected, product yield and quality varied considerably for the different polymerases tested.
The intensity of the product peak is proportional to PCR product yield. Peak intensities reveal yields ranging from poor to good. PCR quality is more difficult to evaluate under non-denaturing conditions than yield.
In one case, however, product quality was apparently not acceptable as evidenced by the presence of multiple spurious peaks, with the target product accounting for only a small portion of the total amount of PCR product amplified. In other cases, only a lack of symmetry of the product peak gives a first indication of potentially poor PCR product quality.
A true measure of PCR product quality is obtained when analyzing amplified PCR products at the software predicted temperature for TMHA and under the software predicted gradient conditions. PCR products that contain polymerase induced errors will form heteroduplexes with the PCR products of the correct sequence.
As seen above, heteroduplexes will elute earlier than their corresponding homoduplexes. Therefore, the extent of UV absorbance recorded preceding the main homoduplex peak is indicative and directly proportional to the amount of PCR product present that carries errors introduced by the polymerase used. In the examples shown in Figure 1-4B all polymerases but Transgenomic’s Optimase Polymerase show significant error incorporation that will interfere with mutation calling.
Heteroduplexes formed due to the presence of sequence variations in the template will co-elute with those heteroduplexes formed due to polymerase-induced errors. As the fidelity of the polymerase used decreases, formation of heteroduplexes resulting from PCR products carrying polymerase-induced errors will increase. As a consequence, accurate and reliable identification of true sequence variations will become increasingly difficult.
In order to maintain high sensitivity and accuracy for mutation detection by DHPLC, or for any other technology for mutation detection, the use of high- fidelity DNA polymerases and PCR conditions promoting polymerase fidelity is essential. Transgenomic recommends the use of Optimase Polymerase for amplification of DNA fragments intended for mutation analysis with the WAVE System.
Conclusion to the Basics of Mutation Screening By DHPLC:
All WAVE Nucleic Acid Fragment Analysis System configurations (2100A, 3500A, 3500HT, and WAVE-MD) provide rapid and high throughput DNA fragment separations on DNASep polymeric cartridges for the purpose of detecting genetic variations such as single nucleotide substitutions, mutations or polymorphisms, as well as short insertions and deletions. Sensitivity and specificity of genetic variance detection with the WAVE System consistently outperforms that of other technologies, including sequencing, at a fraction of their cost and/or labor involved.
The WAVE System provides a highly sensitive, efficient, and inexpensive technology platform for detecting genetic variation. UV and fluorescence on-line detection capabilities afford experimental flexibility. User-friendly software for the prediction of gradient and analysis conditions and fully automated sample handling, analysis, and data acquisition contribute to the high degree of automation and ease of use of the WAVE System technology.