Molecular Genetic Testing in the Genomic Era

Abstract

Technological advancements have allowed molecular-based testing that was once done on a research-only basis to be adopted for routine use in clinical laboratories. Initially, these types of techniques were labor intensive and highly complex. The introduction of automated processes combined with an improved understanding of human genetic variation has allowed molecular testing to expand into clinical diagnostics, where it is now considered an essential aspect of patient care.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Molecular Genetic Testing in the Genomic Era

Technological advancements have allowed molecular-based testing that was once done on a research-only basis to be adopted for routine use in clinical laboratories. Initially, these types of techniques were labor intensive and highly complex. The introduction of automated processes combined with an improved understanding of human genetic variation has allowed molecular testing to expand into clinical diagnostics, where it is now considered an essential aspect of patient care.

Completion of the Human Genome Project has provided the scientific and medical communities with a multitude of genomic targets for diagnostic analysis, including single-nucleotide polymorphisms (SNPs) that have been associated with specific diseases and conditions. Indeed, as a better understanding of genetics and human genetic variation has reached the general public, the demand for genetic testing from both clinicians and patients has increased. Knowledge of the genetic basis of human disease has the potential to affect patient care on multiple levels, including accurate diagnosis, optimal treatment, risk to other family members, as well as prognosis and outcome prediction. The increased use and availability of genetic diagnostic testing has provided improvements to the clinical management of patients. Pharmacogenetic studies on drug metabolism genes have shown that drug responsiveness varies depending on the genetic background of the patient, as in the case of the CYP2C19 polymorphisms and clopidogrel metabolism [1, 2]. Clinical cancer research studies have also shown that the drug responsiveness of some kinds of tumors is often based on specific acquired mutations that can be assayed by molecular testing strategies, as in the case of EGFR mutations in lung cancer and responsiveness to anti-EGFR chemotherapy [3].

In this chapter, we discuss the current state of technology in the clinical diagnostic laboratory setting, with an emphasis on what kinds of testing are best suited for particular medical questions, and also highlight advances in technology that have the potential to transform the field of molecular diagnostics in the near future.

Laboratory Molecular Analysis Methodologies

Nucleic Acid Extraction

The majority of diagnostic tests performed in clinical molecular laboratories are DNA-based, with RNA-based assays making up a distinctly smaller proportion of all testing. While once done solely on a manual basis, extraction of DNA and RNA from biological specimens is typically done using a highly automated process that employs a robotic instrument such as the Qiagen QIAcube or the Roche MagNA Pure System. DNA and RNA can be extracted from different specimen types, though peripheral blood is the most commonly used sample type in most laboratories. For oncology testing, nucleic acid is often isolated from bone marrow or formalin-fixed, paraffin-embedded (FFPE) tumor tissue blocks. The relative instability of RNA requires a fresh or freshly frozen sample for efficient extraction, while DNA can generally be extracted from samples that have been stored for longer periods of time, including FFPE tissue. In many cases, tumors that have been surgically removed from patients several years prior to the availability of genetic tests can be analyzed using current techniques.

Polymerase Chain Reaction

No other method has revolutionized molecular biology, and in turn, molecular diagnostics, to the degree that polymerase chain reaction (PCR) did in the 1980s and 1990s [4], though certainly, next-generation sequencing has the potential to cause another such revolution in the coming years. Until the 1980s, Southern blotting, a time-consuming and labor-intensive technique, was the primary method used by most research and clinical laboratories. The advantages offered by PCR included improved sensitivity and versatility, combined with a reduction in turnaround time for result availability, increasing the utility of molecular testing in a clinical setting.

It is difficult to overemphasize the importance of PCR as it serves as the basis of nearly all genetic diagnostic testing. The discovery of PCR is generally credited to Dr. Kary Mullis, who was awarded the 1993 Nobel Prize in Chemistry in recognition of the improvements he made to a procedure first described many years earlier [5, 6]. Despite its status as an essential technique in molecular biology, PCR is a relatively simple method for generating large amounts of DNA from a small amount of starting material. PCR is a three-step process that is repeated many times to achieve an exponential increase in the quantity of nucleic acid.

To perform basic PCR, a genomic DNA template is combined with two or more oligonucleotide primers, which are small pieces of single-stranded DNA that recognize a specific part of the genome. In an appropriately buffered reaction, DNA polymerase and deoxynucleotide triphosphates (dNTPs), the four nucleotide bases that are needed to make the new strands of DNA, are added to the DNA template and oligonucleotide primers. The three steps of PCR are denaturation, annealing, and extension. In denaturation, the reaction is heated (typically to 94–95°C), and the double-stranded DNA denatures into single strands. In the second step, annealing, the temperature is lowered to approximately 60°C, allowing the oligonucleotide primers to anneal to the complementary sequence in the template DNA. Extension, the final step, occurs at 72°C, the optimal temperature for DNA polymerase. In the extension step, nucleotides are added to form new DNA strands. These three steps are repeated 30–40 times to yield an amplification of the desired piece of DNA, which is often a specific segment of a gene. A representation of the first three cycles of PCR shows the initial exponential amplification from one molecule to eight molecules (Fig. 1.1).

Fig. 1.1

Schematic representation of the first three cycles of a basic PCR amplification. The three steps (denaturation, annealing, and extension) are shown in cycle 1. The target sequence to be amplified is indicated in blue, with flanking sequence in black and primers in red. The products generated during the next two cycles are shown, with eight molecules produced from a single template molecule after three cycles. This exponential amplification generates millions of target molecules after 30 cycles

The discovery of the thermostable Taq polymerase, isolated from Thermus aquaticus, a thermophilic bacterium found in hot springs [7], and the availability of programmable thermocyclers have improved the utility of PCR as a clinical method. Many biotechnology companies offer thermocyclers for traditional PCR, and several companies have developed proprietary reagents that make the PCR process more easily adapted to the clinical laboratory, such as “master mixes” that contain all the reaction components except for the DNA template and the primers.

The ability to generate large amounts of a specific DNA fragment allows for the analysis of many types of genetic variation in human and nonhuman biological specimens. Most laboratory testing methodologies utilize PCR, and it can be adapted for both quantitative and qualitative assays. One of the most powerful aspects of PCR is the incredible versatility of assays that it makes possible, and more recently, adaptations have been introduced that provide even greater flexibility. Some of these variations include allele-specific PCR, multiplex ligation-dependent probe amplification (MLPA), and real-time PCR, all of which will be discussed in subsequent sections of this chapter. The most significant of these modifications is real-time capability, allowing for the simultaneous amplification and detection of nucleic acid targets, eliminating the need for post-PCR processing steps. Real-time PCR has become the method of choice for many laboratories and for many types of tests based on its ability to perform in both qualitative and quantitative applications.

Qualitative Analysis

Most of the clinical testing done in modern molecular genetics laboratories is done to determine the presence or absence of a particular genetic sequence. This qualitative testing can involve genetic variation including inherited and de novo mutations associated with a clinical phenotype, polymorphisms that alter drug metabolism, or acquired somatic alterations associated with the development of neoplasia.

The choice of method for detection of genetic variation is largely dependent on the type of alteration one expects to detect, and whether the changes are frequent and recurrent or unique and highly variable. Certain methods work well for the identification of single-nucleotide changes, epigenetic alterations, and small insertions or deletions, while others are better suited for the detection of large deletions, duplications, or other genomic rearrangements.

Real-Time PCR

Real-time (quantitative) PCR combines traditional PCR with the ability to quantify newly synthesized DNA as it accumulates during each cycle of amplification [8]. The nomenclature for this technique varies, but it is generally accepted that qPCR be used to denote quantitative PCR (real-time PCR), while RT-PCR is reserved for reverse transcription PCR, conversion of RNA to complementary DNA (cDNA), followed by PCR amplification. Real-time PCR is one of the most sensitive tools we have for genetic analysis. The cycle at which amplification crosses a threshold level is directly proportional to the concentration of starting material, allowing for a systematic method of quantification. This is in contrast to traditional PCR, where it is impossible to calculate the starting DNA concentration by measurement of end-point concentration. With qPCR, reporter molecules bind to DNA as it accumulates, which in turn emit a signal that can be measured. The reaction occurs within a chamber that facilitates real-time measurement of the signal, with data uploaded to a computer for recording and plotting. Reporter molecules typically consist of either nonspecific DNA-binding dyes or target-specific primers/probes [9].

A typical PCR reaction starts with reagents in excess of template, which promotes primer-template binding over template renaturation. The exponential phase of PCR is extremely reproducible and is the phase at which data is collected and quantified. The reaction eventually enters a linear phase, which is variable from run to run, followed by cessation of amplification in the plateau phase [8]. In qPCR, the fluorescence of the reaction is measured during each cycle of amplification. More specifically, an “amplification plot” will show the change in fluorescence plotted against cycle number. Since background noise is inevitable, a threshold level of fluorescent signal is set at a point that is determined to be significant (i.e., above background). The cycle at which the measured fluorescence crosses the threshold is called the cycle threshold, or C_t (Fig. 1.2). Normally, amplification is carried out for 35–40 cycles. A logarithmic scale depicts the exponential growth phase as a linear plot which can be used for direct comparison of the amount of DNA to cycle number. Theoretically, the concentration of DNA doubles during every cycle in the exponential phase of a PCR reaction. Therefore, we can calculate the relative concentration of DNA based on the C_t. For example, if sample A has a C_t of 9 and sample B a C_t of 14, then sample A crossed the threshold five cycles earlier than B and has 2^5 (32×) more starting material. In actuality, the reactions are not 100% efficient, so the numbers require some modification, as defined by validation protocols. The specificity of the technique depends largely on good primer design. For example, primer dimers (primers binding to each other rather than target DNA) and nonspecific amplification permit nonspecific dye binding and can result in falsely increased values and possibly false-positive results.

Fig. 1.2

The fold increase over background of fluorescence is plotted on the y-axis, and the cycle number on the x-axis. The point at which the fluorescence crosses the threshold (red horizontal line) is called the Ct. The Ct value is inversely proportional to the concentration of starting DNA. The jagged lines to the left and middle of the graph are nonspecific background noise

DNA-Binding Dyes

DNA-binding dyes provide a nonspecific method of detecting the concentration of double-stranded DNA as it increases with each PCR cycle. Originally, this was done using ethidium bromide, which intercalates in double-stranded DNA [9]. When DNA treated with ethidium bromide is exposed to ultraviolet light (300–360 nm), a measureable orange florescence (590 nm) is emitted. As the concentration of DNA increases, so does the relative amount of intercalated ethidium bromide and, therefore, the fluorescence. SYBR Green, which preferentially binds to the minor groove of double-stranded DNA, is an example of a modern version of this technique [10]. Introduced in the 1990s, SYBR Green allows for a cleaner signal than ethidium bromide, which exhibits background noise from free dye in solution. In addition, modern dyes have been shown to be less mutagenic than ethidium bromide, making them safer alternatives [11].

Target-Specific Methods

Target-specific methods are those that are designed to identify a particular DNA alteration. An important advantage of target-specific methods is that multiplexing is possible. Multiple primers can be labeled with distinguishable fluorescent signals to detect amplification of different targets simultaneously. This technique also produces a more specific signal, reducing background noise and detection of nonspecific amplification products.

An example of a target-specific methodology utilizing qPCR is called TaqMan® [12]. In this method, target-specific PCR primers are created to flank the region of interest. Subsequently, a probe specific for the region of interest hybridizes to the template. The probe contains a fluorescent reporter dye attached to the 5′ end and a quencher moiety coupled to the 3′ end. The proximity of the reporter to the quencher prevents fluorescence emission due to Förster resonance energy transfer (FRET). During the extension phase of PCR, the 5′–3′ exonuclease activity of the Taq DNA polymerase cleaves the 5′ reporter dye, separating the reporter from the quencher, allowing the reporter to fluoresce, with the signal increasing with each successive cycle.

Molecular beacons rely on the same basic principles as TaqMan® probes. The probe itself is a hairpin-shaped molecule, which in the unbound form holds the reporter and quencher in close proximity. The stem of the probe is composed of a sequence that is complementary to the target sequence [13]. Upon hybridization, the hairpin unfolds, separating the reporter from the quencher and allowing for the emission of fluorescence.

Sequencing Analysis

The ability to precisely determine the nucleotide sequence of a particular genomic segment has enabled the detection of many types of genetic variation, including novel mutations that alter gene function, and polymorphic variants that are present in the population at some measurable frequency. Conventional chain-terminator sequencing, also known as Sanger sequencing, after its developer, Frederick Sanger, is commonly used to determine the order of nucleotide bases in a DNA fragment [14]. This type of sequencing utilizes modified nucleotides, known as dideoxy-NTPs (ddNTPs), to terminate chain elongation, resulting in fragments of different lengths that give sequence information for each fragment, based on the incorporated ddNTP. The modified ddNTPs are typically fluorescently labeled, each with a different fluorophore, so all four nucleotide bases of DNA (adenosine, thymine, guanine, and cytosine) can be detected and differentiated. Chain-terminator sequencing has evolved over the past decades to become highly automated, due largely to the use of capillary electrophoresis platforms such as the 3130 Genetic Analyzer and the 3730 DNA Analyzer from Applied Biosystems. Additionally, commercially available dye-terminator reagents such as the BigDye® Terminator Cycle Sequencing kit, also from Applied Biosystems, have made DNA sequencing a method that is both rapid and simple to perform on a clinical basis.

DNA sequencing as a diagnostic tool is most appropriate in situations in which the clinical phenotype is known to be associated with a particular gene, but the alterations can be located in many different segments of the gene. This is exemplified by mutations in the GJB2 gene in nonsyndromic autosomal recessive sensorineural hearing loss [15]. GJB2 encodes the connexin-26 protein, which is expressed in the inner ear and forms ion channels that regulate intercellular communication. Although certain mutations, such as c.35delG (the deletion of a single guanine nucleotide), occur more frequently than others, mutations have been identified throughout the coding region (exon 2) and flanking intronic regions of GJB2. For this reason, sequencing of the coding region is the most efficient method of mutation detection. Sequencing can detect point mutations that change one nucleotide for another, such as the missense mutation c.101T>C [p.M34T], and can also detect small insertion/deletion (indel) mutations, such as c.35delG, that cause a shift in the sequence that is easily visible. Sequencing electropherograms showing these two mutations are in Fig. 1.3.

Fig. 1.3

Sequencing chromatograms showing sequence data from patients that are homozygous normal (top), heterozygous (middle), or homozygous mutant (bottom) for two different GJB2 (connexin 26) mutations. The c.101T>C mutation is shown on the left, and the c.35delG frameshift mutation is shown on the right. The arrows indicate the position of the mutation. Note that the heterozygote for c.101T>C shows two overlapping peaks, one for the T and one for the C. The heterozygous c.35delG chromatogram indicates the frameshifted nature of this mutation, with two sequence traces beginning at the position of the T nucleotide deletion (arrow). The homozygous c.35delG chromatogram shows that both alleles are lacking one of the G nucleotides from the polyG tract

Some tumors are also associated with a spectrum of mutations, such as non-small-cell lung cancer and mutations in the epidermal growth factor receptor (EGFR) gene. Mutations in the kinase domain of EGFR, which spans exons 18 through 21, have been shown to confer responsiveness to tyrosine kinase inhibitors such as gefitinib, erlotinib, and cetuximab [3].

Pyrosequencing

Unlike Sanger sequencing, which relies on “sequencing by termination,” Pyro­sequencing® (Qiagen) relies on a “sequencing by synthesis” methodology [16]. The reaction mixture consists of a template (the sequence to be interrogated), a set of primers, enzymes (DNA polymerase, ATP sulfurylase, luciferase, and apyrase), and substrates (adenosine 5′-phosphosulfate and luciferin). As the reaction progresses, a deoxynucleotide triphosphate (dNTP) is injected into the reaction chamber. If it is complementary to the next base in the template strand, DNA polymerase will incorporate it, releasing one pyrophosphate molecule. ATP sulfurylase will then convert the pyrophosphate to ATP in the presence of adenosine 5′-phosphosulfate. The newly synthesized ATP fuels the luciferase-mediated conversion of luciferin to oxyluciferin, which generates a measureable signal (Fig. 1.4). If the nucleotide is not incorporated, then no signal is produced. Apyrase, a nucleotide-degrading enzyme, continuously degrades unincorporated nucleotides and ATP. When degradation is complete, another nucleotide is injected into the reaction [17]. This series continues sequentially, creating what is called a pyrogram (Fig. 1.5). It has been reported that pyrosequencing has an analytical sensitivity that allows detection of 5% mutant allele; this high sensitivity makes it ideal for situations where tumor cells are admixed with abundant nonneoplastic tissue [18]. For example, solid tumors sent for KRAS or BRAF gene mutation testing typically require dissection of carcinoma from formalin-fixed paraffin-embedded tissue (FFPE) and separation from background stromal cells, making them ideal candidates for pyrosequencing.1.5

Fig. 1.4

The synthesis of ATP drives the conversion of luciferin to oxyluciferin, producing a detectable signal. Apyrase continuously degrades unincorporated nucleotides in preparation for the next dNTP injection

Fig. 1.5

Pyrogram. Addition of dNTPs is performed sequentially. As the process continues, the complementary DNA strand is built up, and the nucleotide sequence is determined from the signal peaks in the pyrogram trace

Promoter Hypermethylation

Alterations in gene expression due to epigenetic changes can have pathological implications similar to the effects of mutations in gene coding sequences. DNA methylation is one mechanism of epigenetic gene silencing. The addition of a methyl group to the carbon-5 of the cytosine pyrimidine ring creates 5-methylcytosine (5-mC). This conversion is an in vivo process which typically occurs in the context of a CpG dinucleotide. CpG islands are often found in gene promoter regions, where methylation (both normal and aberrant) may result in transcriptional silencing of the gene. It has been suggested that methylation has evolved to suppress transcription of regions of the genome consisting of inserted viral sequences, transposons, and repeat elements, all of which may be harmful to a cell if expressed [9]. However, the very same system has been implicated in silencing crucial tumor suppressor genes in various carcinomas, such as the VHL gene in renal carcinoma or the MLH1 gene in colorectal and other carcinomas [19].

There are examples of other types of genes, such as repair enzymes, being silenced through methylation which confer an advantage under certain circumstances. Recently, methylation of the promoter of the O⁶-methylguanine methyltransferase (MGMT) gene has been reported in some cases of glioblastoma and is associated with prolonged survival in patients treated with temozolomide [20]. MGMT is a DNA repair enzyme that reverses alkylation of guanine by transferring the alkyl group to the active site of the enzyme. Diminished MGMT expression due to methylation of CpG sites in the 5′ region of the MGMT gene allows accumulation of alkylguanine DNA which, following incorrect base pairing with thymine, triggers DNA damage signaling and cell death. MGMT methylation is associated with increased sensitivity of glioblastoma to alkylating agents such as BCNU (carmustine), and it also correlates with prolonged progression-free survival in glioma patients treated with temozolomide [21].

In vitro, treatment of DNA with bisulfite converts unmethylated cytosine residues to uracil (Fig. 1.6), while methylated nucleotides are protected from this modification [22]. By sequencing the promoter region of a gene both before and after bisulfite treatment, one can calculate the percent methylation to determine if a gene is transcriptionally silent (Fig. 1.7).

Fig. 1.6

Bisulfite conversion of cytosine to uracil. Reaction 1: The reversible addition of bisulfite (HSO3−) results in an equilibrium between C and C-SO3−. Acidic conditions shift the equilibrium toward C-SO3− side. Reaction 2: The deamination of C-SO3− is the rate-limiting step, and the velocity increases in proportion to bisulfite concentration. Reaction 3: The final step results in the conversion of U-SO3− to U and recovery of the 5,6-double bond

Fig. 1.7

Pyrograms of two different patient samples are shown (top and bottom panes) after bisulfite treatment. The sequence of the region interrogated is: CGA CAG CGA TAC GCG, with cytosines of interest underlined and what is called the “natural control” position italicized. Note: U residues are detected as T residues in DNA sequencing. Top – All C residues (blue shading) were converted, that is, the segment of DNA is not methylated; bottom – first position – 57% did not convert (i.e., 57% remained as a cytosine). The next position shows the natural control (tan shading), where the C residue fully converted as expected. (It is not a CpG dinucleotide; therefore, it is never methylated and always converts.) The following three positions are calculated the same way. The average of the four positions (not including the natural control) is calculated at 58% methylated, enough to consider the gene silenced (the actual cutoff value is determined during validation procedures in the lab)

Fragment Analysis

In situations in which the genetic alteration affects the size of the DNA fragment, either by the removal or addition of nucleotides, fragment analysis is the simplest method of detection. Conveniently, the same capillary electrophoresis platforms used for DNA sequencing can also be used for DNA fragment analysis. An example of this methodology is the detection of mutations in two different genes commonly mutated in patients with acute myelogenous leukemia (AML) with normal cytogenetics: FLT3 and NPM1 [23, 24]. In the FLT3 gene, the most commonly observed alteration is an in-frame internal tandem duplication (ITD) within the coding sequence of the juxtamembrane domain found in exons 13 and 14. This mutation results in the constitutive activation of the FLT3 tyrosine kinase, and its presence confers a high risk of relapse and an overall poorer prognosis than that observed in AML patients that do not harbor this mutation. In the NPM1 gene, the most common mutation observed in AML patients is a 4-bp insertion in exon 12 of the gene that interrupts the gene’s nuclear localization signal and results in the aberrant cytoplasmic localization of the nucleophosmin protein. This mutation is associated with an overall favorable prognosis in the absence of a concomitant FLT3 ITD mutation, and an intermediate prognosis if the FLT3 ITD is present. Examples of fragment analysis of PCR products showing amplification of FLT3 from 3 AML samples with different sizes of the ITD mutation are shown in Fig. 1.8a. A peak representing the normal-sized allele is present as well as a peak representing a larger mutant fragment in all three samples. Figure 1.8b shows amplification of the NPM1 gene in the area of the four-base-pair insertion. Two samples are positive for the mutation, and one does not have the mutation present (bottom panel).

Fig. 1.8

Fragment analysis results for two tests done for AML. In a, three patients with different FLT3 ITD mutations are shown. The normal-sized peak is the same in all three, and different-sized mutant peaks are present in each sample, indicating that this mutation is of a variable size. In b, the results for NPM1 mutation analysis are shown for three patients. All three show the normal peak. The top two panels are from patients that are positive for the NPM1 four-base-pair insertion, visible as a second, larger peak. The bottom panel is a sample from a patient lacking the NPM1 mutation and only shows a single peak. The less prominent peaks surrounding the high peaks are commonly observed in fragment analysis and represent PCR products that differ by one base in size and are generally due to slippage during PCR

Multiplex PCR and Fragment Analysis

Multiplex PCR is the process whereby multiple sets of PCR primers are utilized to amplify several areas of interest in a single PCR reaction. To differentiate the amplicons, each primer set can be labeled with a unique fluorescent tag, or the amplicons can be of different lengths such that they can be differentiated by electrophoresis. One example of the use of multiplex PCR in the clinical laboratory is in microsatellite instability (MSI) analysis. Microsatellites, tracts of short repeated elements that typically range from 1 to 7 nucleotides in length, vary in length between individuals. However, in patients with defects in mismatch repair genes (e.g., MLH1, MSH2, MSH6 and PMS2), variation in the length of microsatellites can vary between normal tissue and that taken from a tumor sample in the same individual. MSI analysis is able to detect these differences by amplifying the same microsatellites in both normal and tumor tissue and then visualizing the results by capillary electrophoresis (or an equivalent method). The detection of MSI can suggest the diagnosis of Lynch syndrome (also known as hereditary nonpolyposis colon cancer or HNPCC), a dominantly inherited cancer predisposition syndrome associated with an increased risk for multiple cancer types including colorectal, endometrial, ovarian, urothelial, and small intestine cancers. One of the hallmarks of Lynch syndrome–associated tumors is an increase in DNA replication errors, and MSI, due to mismatch repair deficiency. Microsatellites used in clinical testing are typically mononucleotide repeat sequences, and in tumors with mismatch repair defects, the errors are visible as the presence of new alleles in the tumor sample. If a patient is found to have MSI in at least 2 of the 5 microsatellites tested, it is considered MSI high. For patients with MSI-high tumors, further testing for Lynch syndrome can be performed. A diagnosis of Lynch syndrome in a family can have a dramatic impact on cancer risk estimates for the other members of the family that have inherited the gene defect and result in increased screening with the goal of early detection. Figure 1.9 shows the typical results in an MSI-high tumor and a microsatellite stable tumor.

Fig. 1.9

Microsatellite instability analysis of matched tumor and normal tissue from two patients. Panel A is the normal, and B is the tumor from a patient with colorectal cancer, and the tumor panel (B) shows high microsatellite instability, with shifts in microsatellite size in all five markers examined (arrows indicate new alleles not present in normal tissue). Panels C and D are from a patient with a colorectal cancer that is microsatellite stable, and no new alleles are present in the tumor (D)

Melting Curve Analysis

Sanger sequencing of DNA is the gold standard for identifying germline mutations; however, it is time consuming and labor intensive. As such, methods have been developed to “scan” DNA to quickly determine if a region of interest is likely to contain a mutation. One such method is high-resolution melt analysis (HRM) [25]. HRM is performed by monitoring the melting temperature (T_m) of double-stranded DNA (called duplexes) during controlled heating. Homoduplexes are strands of DNA that are perfectly complementary to one another, whereas heteroduplexes contain at least one mismatched base pair. These duplexes create a melt curve signature over a range of temperatures. The process is usually carried out in a light cycler following PCR. Melting curve analysis exploits the fact that mismatched strands of DNA (even at a single nucleotide) will have a lower T_m than perfectly complementary strands. The simplest method to monitor this reaction involves the use of a double-stranded DNA-binding dye at saturating concentrations. Figure 1.10 shows a shift in T_m that is attributed to a single base change of a pyrimidine to a purine between 2 homoduplexes, normal (wild-type) and mutant. As with most screening tools, possibly significant results identified using HRM are investigated further by another more specific method (i.e., DNA sequencing).

Fig. 1.10

A pyrimidine to purine base change in a segment of DNA will change the temperature at which the strands denature or “melt.” The change in T_m is noted as a shift on the graph. The shift represents denaturing of double-stranded DNA at a different temperature and reduced signal emission from the double-stranded DNA-binding dye

HRM can be used to screen for a wide range of mutation types but is often used for the detection of point mutations or single-nucleotide polymorphisms (SNPs). If a patient’s DNA is mixed with a known sequence of DNA, HRM can determine if there are mismatched nucleotides (heteroduplexes) by comparing the melt curve to that of known heteroduplexes or homoduplexes (Fig. 1.11).

Fig. 1.11

There are three scenarios: wild-type, heterozygous, or homozygous. Each gives a melt curve that is slightly different. With a high-quality HRM assay, it is possible to distinguish between all three of these scenarios

Multiplex Ligation-Dependent Probe Amplification

One of the variations of PCR that can effectively and easily detect large genomic deletions and duplications that affect one or more whole exons of a gene is multiplex ligation-dependent probe amplification (MLPA) [26]. Large deletions/duplications comprise a significant proportion of the mutations responsible for some genetic conditions, and MLPA provides a rapid, reliable, and inexpensive way to detect them. For example, Duchenne muscular dystrophy (DMD) is a severe X-linked recessive form of muscular dystrophy caused by mutations of the dystrophin (DMD) gene. Often, the causative mutations in patients with DMD are large-scale genomic alterations, including deletion (∼65%) and duplication (∼7–10%) of one or more exons. The remaining ∼25% of cases are associated with point mutations or small insertions and deletions that can be detected by sequence-based methods [27].

MLPA has advantages over traditional multiplex PCR in that only a single PCR primer pair that amplifies multiple probes hybridized to the tested gene. The MLPA probes consist of two separate oligonucleotides, each containing one of the PCR primer sequences. It is only when these two hemiprobes are both hybridized to their adjacent targets and ligated that they can be amplified by PCR in a quantitative fashion (Fig. 1.12, reproduced with permission from MRC-Holland). For a diagnostic case, a patient sample is compared to a reference sample, and copy number changes can be assessed for each target sequence (usually an individual exon). Conveniently, the biotechnology company MRC-Holland has developed many commercially available kits for MLPA that are commonly used in clinical laboratories [28].

Fig. 1.12

Outline of the MLPA® technique. After hybridization to their target sequence in the sample DNA, the probe oligonucleotides are enzymatically ligated. One probe oligonucleotide contains a nonhybridizing stuffer sequence of variable length. Ligation products can be amplified using PCR primer sequences X and Y; amplification product of each probe has a unique length (130–500 nt). Amplification products are separated by electrophoresis. Relative amounts of probe amplification products, as compared to a reference DNA sample, reflect the relative copy number of target sequences

The Next Generation of Molecular Diagnostics

The increased demand for low-cost, high-throughput sequencing has led to advances in technology in recent years that are likely to dramatically alter the landscape of clinical molecular diagnostic testing in the same way PCR transformed the field of molecular biology over two decades ago. The term “next-generation sequencing” (NGS) refers broadly to platforms that allow massively parallel DNA sequencing. The different technologies utilized are all fundamentally different than traditional Sanger sequencing [29]. Unlike Sanger sequencing, which generates a consensus sequence from a pool of DNA molecules, NGS methods produce individual sequence reads from a single strand of DNA, and on average, each nucleotide is sequenced many times (typically 20–60 times). The sequences are subsequently combined using computational methods. NGS has many advantages over traditional sequencing methods (reduced cost per sequenced nucleotide, potential for automation, and a wider variety of alterations that can be identified) that will ultimately result in its widespread adoption in clinical diagnostics laboratories. Next-generation sequencing has the potential to replace many of the currently utilized diagnostic tools due to its ability to detect large-scale genomic rearrangements including translocations, inversions, deletions, and duplications in addition to single-nucleotide alterations and small insertions and deletions.

Clinical Applications of Next-Generation Sequencing

The potential clinical uses of next-generation sequencing include sequencing of the whole genome, exome (all gene coding exons), or transcriptome (all expressed sequences). To date, these applications have primarily been used to identify novel genes associated with genetic disorders in basic research and translational studies. However, several clinical laboratories have started to use this technology as part of patient care, and it will likely become the method of choice for most laboratories in the very near future.