Molecular Microbiology

Since the publication of the ninth edition of this Manual, significant changes have occurred inthe practice of diagnostic molecular microbiology. Nucleic acid amplification techniques arecommonly used to diagnose and manage patients with infectious diseases. The growth in thenumber of commercially available test kits and analyte-specific reagents (ASRs) hasfacilitated the use of this technology in the clinical laboratory. Technological advances inreal-time PCR techniques, automation, nucleic acid sequencing, multiplex analysis, and massspectrometry have reinvigorated the field and created new opportunities forgrowth.

NONAMPLIFIED NUCLEIC ACID PROBES

Nucleic acid probes are segments of DNA or RNA labeled with radioisotopes, enzymes, orchemiluminescent reporter molecules that can bind to complementary nucleic acid sequenceswith high degrees of specificity. Although probes can range from 15 to thousands ofnucleotides in size, synthetic oligonucleotides of less than 50 nucleotides are most commonlyincorporated into commercial kits. The probes can be designed to identify microorganisms atany taxonomic level. A number of commercially available DNA probes have been developedfor direct detection of pathogens in clinical specimens and identification of pathogens afterisolation by culture.The commonly used formats for probe hybridization include liquid-phase, solid-phase, and insitu hybridization. The leading method used in clinical microbiology laboratories is a liquidphasehybridization protection assay (Gen-Probe, Inc., San Diego, CA). In this method, asingle-stranded DNA (ssDNA) probe labeled with an acridinium ester is incubated with thetarget nucleic acid. Alkaline hydrolysis follows the hybridization step, and probe binding ismeasured in a luminometer after the addition of peroxides. For a positive sample, theacridinium ester on the bound probe is protected from hydrolysis and, upon the addition of

peroxides, emits light. The hybridization protection assay can be completed in several hoursand does not require removal of unbound single-stranded probe or isolation of probe-bounddouble-stranded sequences (3).

In solid-phase hybridization, target nucleic acids are bound to nylon or nitrocellulose and arehybridized with a probe in solution (164). The unbound probe is washed away, and thebound probe is detected by means of fluorescence, luminescence, radioactivity, or colordevelopment. Although solid-phase hybridization is a powerful research tool, the length oftime required and the complexity of the procedure limit its application in clinical practice.

In situ hybridization is another type of solid-phase hybridization in which the nucleic acid iscontained in tissues or cells which are affixed to microscope slides and is governed by thesame basic principles as described previously (55). In most clinical applications, formalinfixed,paraffin-embedded tissue sections are used. The sensitivity of in situ hybridization isoften limited by the accessibility of the target nucleic acid in the cells.

In general, due to the poor analytical sensitivities of nonamplified-probe techniques, theapplication of these techniques to direct detection of pathogens in clinical specimens is limited to those situations in which the number of organisms is large. Such situations includecases of group A streptococcal pharyngitis, genital tract infections with Neisseria gonorrhoeae and Chlamydia trachomatis, and agents associated with vaginosis and vaginitis.

These techniques are used most effectively in culture confirmation assays for mycobacteria and systemic dimorphic fungi. These culture confirmation tests have a positive effect on patient management by providing rapid and accurate detection of these slowly growing, often difficult-to-identify pathogens.

Nucleic acid probes for direct detection of group A streptococci, C. trachomatis, and N. gonorrhoeae are available from Gen-Probe. Probes for identification of Blastomyces dermatitidis, Coccidioides immitis, Histoplasma capsulatum, campylobacters, enterococci, group A streptococci, group B streptococci, Haemophilus influenzae, Listeria monocytogenes, mycobacteria, N. gonorrhoeae, Staphylococcus aureus, and Streptococcus pneumoniaeisolated in culture are also available from Gen-Probe.

A solid-phase nucleic acid probe test for detection and identification of Gardnerella vaginalis, Trichomonas vaginalis, and Candida albicans in vaginal fluid from patients with vaginosis or vaginitis is available from BD Diagnostic Systems, Sparks, MD. It uses two distinct probes for each organism, a capture probe and a color development probe, in an easy-to-use format.

Peptide nucleic acid (PNA) probes are DNA mimics in which the negatively charged sugar

phosphate backbone of DNA is replaced with a noncharged polyamide or “peptide” backbone. PNA probes contain the same nucleotide bases as DNA and follow standard Watson-Crick base pairing rules when hybridizing to complementary nucleic acid sequences (153). Because PNA probes are noncharged, they do not have to overcome the destabilizing electrostatic

repulsion that occurs when two negatively charged DNA molecules hybridize. As a result, PNA probes bind more rapidly and tightly to nucleic acid targets. In addition, the relatively hydrophobic character of the PNA probes enables them to penetrate the hydrophobic cell membrane following preparation of a standard smear. PNA probes have been used for identification of S. aureus, Escherichia coli, Pseudomonas aeruginosa, andCandida albicans directly from positive blood culture bottles (120, 135, 150) and direct detection ofMycobacterium tuberculosis in smear-positive sputum samples (154).

PNA probes for rapid, direct identification of S. aureus, coagulase-negative

staphylococci, Enterococcus faecalis, Escherichia coli, Pseudomonas

aeruginosa,and Candida spp. from positive blood culture bottles and Streptococcus

agalactiae from Lim broth cultures are available from AdvanDx, Woburn, MA

(106, 145, 150).

TECHNIQUES USING AMPLIFIED NUCLEIC ACIDS Back to top

The development of the PCR by Saiki et al. (140) was a milestone in biotechnology and heralded the beginning of molecular diagnostics. PCR had its 20th birthday in 2005 and has

stood the test of time. Although PCR is the best-developed and most widely used nucleic acid amplification strategy, other strategies have been developed, and several have clinical utility. These strategies are based on signal, target, or probe amplification. Examples of each category are discussed in the sections that follow. These techniques have sensitivity unparalleled in laboratory medicine, have created new opportunities for the clinical laboratory to have an effect on patient care, and have become the new “gold standards” for laboratory diagnosis of many infectious diseases.

SIGNAL AMPLIFICATION TECHNIQUES

In signal amplification methods, the concentration of the probe or target does not increase. The increased analytical sensitivity comes from increasing the concentration of labeled

molecules attached to the target nucleic acid. Multiple enzymes, multiple probes, multiple layers of probes, and reduction of background noise have all been used to enhance target detection (74). Target amplification systems generally have greater analytical sensitivity than signal amplification methods, but technological developments, particularly in branched-DNA (bDNA) assays, have lowered the limits of detection to levels that may rival those of

target amplification assays in some applications (69).

Signal amplification assays have several advantages over target amplification assays. In signal amplification systems, the number of target molecules is not altered, and as a result, the signal is directly proportional to the amount of the target sequence present in the clinical specimen. This reduces concerns about false-positive results due to cross contamination and

simplifies the development of quantitative assays. Since signal amplification systems are not dependent on enzymatic processes to amplify the target sequence, they are not affected by the presence of enzyme inhibitors in clinical specimens. Consequently, less cumbersome nucleic acid extraction methods may be used. Typically, signal amplification systems use

either larger probes or more probes than target amplification systems and, consequently, are less susceptible to errors resulting from target sequence heterogeneity. Finally, RNA levels

can be measured directly without the synthesis of a cDNA intermediate.

bDNA Assays

The bDNA signal amplification system is a solid-phase, sandwich hybridization assay incorporating multiple sets of synthetic oligonucleotide probes (114). The key to this

technology is the amplifier molecule, a bDNA molecule with 15 identical branches, each of which can bind to three labeled probes.

The bDNA signal amplification system is illustrated in Fig. Fig. 1. Multiple target-specific probes are used to capture the target nucleic acid onto the surface of a microtiter well. A second set of target-specific probes also binds to the target. Preamplifier molecules bind to the second set of target probes and up to eight bDNA amplifiers. Three alkaline phosphataselabeled probes hybridize to each branch of the amplifier. Detection of bound labeled probes is

achieved by incubating the complex with dioxetane, an enzyme-triggerable substrate, and measuring the light emission in a luminometer. The resulting signal is directly proportional to the quantity of the target in the sample. The quantity of the target in the sample is determined from an external standard curve.

Nonspecific hybridization of any of the amplification probes and nontarget nucleic acids leads

to amplification of the background signal. In order to reduce potential hybridization to

nontarget nucleic acids, isocytidine (isoC) and isoguanosine (isoG) were incorporated into the

preamplifier and labeled probes were used in the third-generation bDNA assays (23). IsoC

and isoG form base pairs with each other but not with any of the four naturally occurring

bases (130).

The use of isoC- and isoG-containing probes in bDNA assays increases target-specific signal

amplification without a concomitant increase in the background signal, thereby greatly

enhancing the detection limits without loss of specificity. The detection limit of the thirdgeneration

bDNA assay for human immunodeficiency virus type 1 (HIV-1) RNA is 75

copies/ml. bDNA assays for the quantification of hepatitis B virus (HBV) DNA, hepatitis C

virus (HCV) RNA, and HIV-1 RNA are commercially available (Siemens Healthcare

Diagnostics, Deerfield, IL). The System 340 and 440 analyzers for bDNA assays automate

the incubation, washing, reading, and data processing steps.

Hybrid Capture Assays

The hybrid capture system is a solution hybridization- antibody capture method that uses

chemiluminescence detection of the hybrid molecules. The target DNA in the specimen is

denatured and then hybridized with a specific RNA probe. The DNA-RNA hybrids are captured

by antihybrid antibodies that are used to coat the surface of a tube. Alkaline phosphataseconjugated

antihybrid antibodies bind to the immobilized hybrids. The bound antibody

conjugate is detected with a chemiluminescent substrate, and the light emitted is measured

in a luminometer. Multiple alkaline phosphatase conjugates bind to each hybrid molecule,

amplifying the signal. The intensity of the emitted light is proportional to the amount of

target DNA in the specimen. Hybrid capture assays for detection of N. g onorrhoeae, C.

trachomatis, human papillomavirus (HPV) (25), and cytomegalovirus (CMV) (102) in clinical

specimens have been developed (Qiagen, Germantown, MD).

TARGET AMPLIFICATION TECHNIQUES

All of the target amplification systems share certain fundamental characteristics. They use

enzyme-mediated processes, in which a single enzyme or multiple enzymes synthesize

copies of target nucleic acid. In all of these techniques, the amplification products are

detected by two oligonucleotide primers that bind to complementary sequences on opposite

strands of double-stranded targets. All the techniques result in the production of millions to

billions of copies of the targeted sequence in a matter of hours, and in each case, the

amplification products can serve as templates for subsequent rounds of amplification.

Because of this, all of the techniques are sensitive to contamination with product molecules

that can lead to false-positive reactions. The potential for cross contamination is real and

should be adequately addressed before any of these techniques are used in the clinical

laboratory. However, the occurrence of false-positive reactions can be reduced through

special laboratory design, practices, and work flow.

Polymerase Chain Reaction

PCR is a simple, in vitro, chemical reaction that permits the synthesis of essentially limitless

quantities of a targeted nucleic acid sequence. This is accomplished through the action of a

DNA polymerase that, under the proper conditions, can copy a DNA strand (Fig. 2). At its

simplest, a PCR consists of target DNA, a molar excess of two oligonucleotide primers, a

heat-stable DNA polymerase, an equimolar mixture of deoxyribonucleotide triphosphates

(dNTPs; dATP, dCTP, dGTP, and dTTP), MgCl2, KCl, and a Tris-HCl buffer. The two primers

flank the double-stranded DNA (dsDNA) sequence to be amplified, typically <100 to several

hundred bases, and are complementary to opposite strands of the target.

To initiate a PCR, the reaction mixture is heated to separate the two strands of target DNA

and is then cooled to permit the primers to anneal to the target DNA in a sequence-specific

manner. The DNA polymerase then initiates extension of the primers at their 3′ ends toward

one another. The primer extension products are dissociated from the target DNA by heating.

Each extension product, as well as the original target, can serve as a template for

subsequent rounds of primer annealing and extension.

At the end of each cycle, the PCR products are theoretically doubled. Thus, after n PCR

cycles the target sequence can be amplified 2n-fold. The whole procedure is carried out in a

programmable thermal cycler that precisely controls the temperature at which the steps

occur, the lengths of time that the reaction mixture is held at the different temperatures, and

the number of cycles. Ideally, after 20 cycles of PCR a 106-fold amplification is achieved and

after 30 cycles a 109-fold amplification occurs. In practice, the amplification may not be

completely efficient due to failure to optimize the reaction conditions or the presence of

inhibitors of the DNA polymerase. In such cases, the total amplification is best described by

the expression (1 + e)n, where e is the amplification efficiency (0 ≤ e ≤ 1) and n is the total

number of cycles.

Reverse Transcriptase PCR

As it was originally described, PCR was a technique for DNA amplification. Reverse

transcriptase PCR (RT-PCR) was developed to amplify RNA targets. In this process, cDNA is

first produced from RNA targets by reverse transcription and then the cDNA is amplified by

PCR. As it was originally described, RT-PCR used two enzymes: a heat-labile RT, such as

avian myeloblastosis virus RT, and a thermostable DNA polymerase. Because of the

temperature requirements of the heat-labile enzyme, cDNA synthesis had to occur at

temperatures below the optimal annealing temperatures of the primers. This presented

problems in terms of both nonspecific primer annealing and inefficient primer extension due

to the formation of RNA secondary structures. These problems have largely been overcome

by the development of a thermostable DNA polymerase derived from Thermus

thermophilus that under the proper conditions can function efficiently as both an RT and a

DNA polymerase (109). RT-PCRs with this enzyme are more specific and efficient than

previous protocols with conventional, heat-labile RT enzymes.

Nested PCR

Nested PCR was developed to increase both the sensitivity and the specificity of PCR (56). It

uses two pairs of amplification primers and two rounds of PCR. Typically, one primer pair is

used in the first round of PCR for 15 to 30 cycles. The products of the first round of

amplification are then subjected to a second round of amplification with the second set of

primers, which anneal to a sequence internal to the sequence amplified by the first primer

set. The increased sensitivity arises from the high total cycle number, and the increased

specificity arises from the annealing of the second primer set to sequences found only in the

first-round products, thus verifying the identity of the first-round product. The major

disadvantage of nested PCR is the high rates of contamination that can occur during the

transfer of first-round products to the second tube for the second round of amplification. This

contamination can be avoided either by physically separating the first- and second-round

amplification mixtures with a layer of wax or oil or by designing single-tube amplification

protocols. In practice, the enhanced sensitivity afforded by nested PCR protocols is rarely

required in diagnostic applications, and the identity of an amplification product is usually

confirmed by hybridization with a nucleic acid probe.

Multiplex PCR

In multiplex PCR, two or more primer sets designed for amplification of different targets are

included in the same reaction mixture (13). By this technique, more than one target

sequence in a clinical specimen can be coamplified in a single tube. The primers used in

multiplexed reactions must be carefully selected so that they have similar annealing

temperatures and lack complementarity. Multiplex PCRs have proved to be more complicated

to develop and may be less sensitive than PCRs with single primer sets.

Many multiplex assays have been developed, especially for the detection of central nervous

system (8, 26) and respiratory (70, 162) pathogens. Multiplex PCR assays for bacterial and

viral respiratory pathogens are commercially available from Prodesse, Inc., Waukesha, WI.

A promising new platform for multiplex PCR analysis is the xMAP system (Luminex Corp.,

Austin, TX). The xMAP system incorporates a proprietary process to internally dye

polystyrene microspheres with two spectrally distinct fluorochromes. By using precise ratios

of these fluorochromes, an array is created consisting of 100 different microsphere sets with

specific spectral addresses. Each microsphere set can possess a different reactant on its

surface. For nucleic acid analysis, oligonucleotide probes would be covalently bound to the

microsphere surface by carbodiimide coupling. Since each microsphere set can be

distinguished by its spectral address, the sets can be combined, allowing up to 100 different

analytes to be measured simultaneously in a single reaction vessel. A third fluorochrome

coupled to a reporter molecule quantifies the biomolecular interaction that occurs at the

microsphere surface.

Microspheres are interrogated individually in a rapidly flowing liquid stream as they pass by

two separate lasers in the Luminex xMAP flow cytometer. High-speed digital signal

processing classifies each microsphere based on its spectral address and quantifies the

reaction on its surface. Thousands of microspheres are investigated per second, resulting in

an analysis system capable of analyzing and reporting up to 100 different reactions in a

single reaction vessel in a few seconds.

Multiplex assays run on the Luminex platform typically consist of three major steps: nucleic

acid amplification by PCR, target-specific extension, and liquid bead array decoding. After

PCR amplification, the amplicons are mixed with a second set of tagged primers specific for

each target. If the target is present, the tagged primer will be extended through a process

called target-specific extension. During this extension, a label is incorporated into the

extension product. The color-coded beads are added to identify the tagged and labeled

extension products. Attached to each differently colored bead is oligonucleotide

complementary to the tag sequence for each target. Samples are then placed in the Luminex

xMAP flow cytometer, where the beads are read by two color lasers. One laser identifies the

color of the bead, and the other laser detects the presence or absence of a labeled extension

product on that bead.

The technology has been adapted to a wide variety of applications in bacteriology (30),

mycology (27), and virology (148, 175). Systems for the multiplex detection of respiratory

viruses based on the Luminex xMAP system have been developed by Luminex Molecular

Diagnostics, EraGen Biosciences (Madison, WI), and Qiagen (10, 98, 113).

Another promising technology for high-order multiplex PCR is the FilmArray, developed by

Idaho Technology, Salt Lake City, UT. It is a completely automated, integrated, and selfcontained

lab-in-a-pouch system. The film portion of the pouch has stations for cell lysis,

nucleic acid purification, reverse transcription to detect RNA targets, first-stage PCR

multiplex PCR, and an array of up to 120 second-stage nested PCRs. After extracting and

purifying nucleic acids from the unprocessed sample, the FilmArray performs a nested

multiplex PCR that is executed in two stages. During the first-stage PCR, the FilmArray

performs a single, large-volume, massively multiplexed reaction. The products from firststage

PCR are then diluted and combined with a fresh, primer-free master mix. Aliquots of

this second master mix solution are then distributed to each well of the array. Each well of

the array is prespotted with a single set of primers. The second-stage, small-volume PCR is

performed in singleplex fashion in each well of the array. Though this assay uses nested PCR,

the entire test is performed within a sealed pouch, thus eliminating concerns of carryover

contamination. Using amplification and melting-curve data, the FilmArray software

automatically generates a result for each target. A FilmArray for detection of 20 different

respiratory pathogens is in development.

Real-Time (Homogeneous, Kinetic) PCR

The term real-time PCR refers to methods in which the target amplification and detection

steps occur simultaneously in the same tube (homogeneous). These methods require special

thermal cyclers with precision optics that can monitor the fluorescence emission from the

sample wells. The computer software supporting the thermal cycler monitors the data

throughout the PCR at every cycle and generates an amplification plot for each reaction

(kinetic).

Figure 3 shows a representative amplification plot and defines the terms used in quantitative

real-time PCR. The amplification plot shows the normalized fluorescence signal from the

reporter at each cycle number. In the initial cycles of PCR, there is little change in the

fluorescence signal. This initial signal level defines the baseline for the plot. An increase

above the baseline indicates the detection of accumulated PCR product. A fixed fluorescence

threshold can be set above the baseline. The cycle threshold (CT) is defined as the cycle

number at which the fluorescence passes the fixed threshold. A plot of the log of the initial

target concentration versusCT for a set of standards is a straight line (59). The amount of the

target in an unknown sample is determined by measuring the sample CT and using a

standard curve to determine the starting copy number. Alternatively, the cycle number

corresponding to the maximal change in fluorescence, the second derivative maximum, has a

similar relationship to the initial target concentration.

In its simplest format, the PCR product is detected as it is produced by using fluorescent

dyes that preferentially bind to dsDNA. SYBR green I is one such dye that has been used in

this application (107). In the dye’s unbound state, the fluorescence is relatively low, but

when the dye is bound to dsDNA, the fluorescence is greatly enhanced. The dye binds to

both specific and nonspecific PCR products. The specificity of the detection can be improved

through melting-curve analysis. As the temperature is slowly raised, the two strands of the

amplicon melt apart and the amount of fluorescence decreases. The data are transformed

and analyzed by plotting the first derivative of the fluorescence on the y axis and the

temperature on the x axis. The specific amplified product will have a characteristic melting

peak at its predicted melting temperature (Tm), whereas the primer dimers and other

nonspecific products should have different Tms or give broader peaks (136).

The specificity of real-time PCR can also be increased by including fluorescent resonance

energy transfer (FRET) probes in the reaction mixture. These probes are labeled with

fluorescent dyes or with combinations of fluorescent and quencher dyes. In 5′ exonuclease

PCR (TaqMan) assays, the 5′- to-3′ exonuclease activity ofTaq DNA polymerase is used to

cleave a nonextendable hybridization probe during the primer extension phase of PCR (61).

This approach uses dually labeled fluorogenic hybridization probes and is illustrated in

Fig. Fig. 4. One fluorescent dye serves as a reporter, and its emission spectrum is quenched

by the second fluorescent dye. The nuclease degradation of the hybridization probe releases

the reporter dye, resulting in an increase in the peak fluorescent emission. The increase in

fluorescent emission indicates that specific PCR product has been made, and the intensity of

fluorescence is related to the amount of the product (57). The specificity is increased

because a signal is generated only when the primer and probe are bound to the same

template strand.

The use of dual hybridization probes is another approach to real-time PCR (81). This method

uses two specially designed sequence-specific oligonucleotide probes (Fig. 5). These

hybridization probes are designed to hybridize within 1 to 5 nucleotides apart on the product

molecule. The 3′ end of the anchor probe is labeled with a donor dye, and the 5′ end of the

reporter probe is labeled with an acceptor dye. The 3′ end of the reporter probe is

phosphorylated to prevent extension during PCR. The donor dye is excited by an external

light source, and instead of emitting light, it transfers its energy to the acceptor dye by

FRET. The excited acceptor dye emits light at a longer wavelength than the unbound donor

dye, and the intensity of the acceptor dye light emission is proportional to the amount of PCR

product.

Real-time detection and quantification of amplification products can also be accomplished

with molecular beacons (171). Molecular beacons are hairpin-shaped oligonucleotide probes

with an internally quenched fluorophore whose fluorescence is restored when the probes bind

to a target nucleic acid (Fig. 6). The probes are designed in such a way that the loop portion

of each probe molecule is complementary to the target sequence. The stem is formed by the

annealing of complementary arm sequences on the ends of the probe. A fluorescent dye is

attached to one end of one arm, and a quenching molecule is attached to the end of the

other arm. The stem keeps the fluorophore and quencher in close proximity such that no

light emission occurs. When the probe encounters a target molecule, it forms a hybrid that is

longer and more stable than the stem and undergoes a conformational change that forces

the stem apart, causing the fluorophore and the quencher to move away from each other,

restoring the fluorescence.

Scorpion probes combine a PCR primer with a molecular beacon (167, 180). Intramolecular

hybridization of the loop structure to a downstream portion of the amplification product

separates the reporter and quencher dyes. The hybridization kinetics of Scorpion probes are

generally faster than those of molecular beacons because the primer and probe are located

on the same molecule.

Dark quencher probes are also used in real-time PCR applications (Epoch Biosciences,

Bothell, WA). Dark quencher probes contain a fluorophore on the 5′ end and a

nonfluorescent quencher molecule on the 3′ end (78). The fluorescence is quenched when

the probe is a random coil and emitted when the probe anneals to the target sequence.

Unlike fluorogenic 5′ nuclease probes, these probes are not degraded by the DNA

polymerase during target amplification. Since the dark quencher is not fluorescent, it does

not contribute to the background signal. This trait has the advantage of improving the signalto-

noise ratio for the detection system, which may improve sensitivity. These probes also

incorporate a hybridization-stabilizing compound, known as a minor groove binder. It is a

small, crescent-shaped molecule that is covalently linked to the 3′ end of the probe that

spans about 3 or 4 nucleotides and snugly fits into the minor groove of DNA, where it forms

hydrogen bonds with the template. Minor groove binders increase the Tm of the probe. The

minor groove binder allows for the use of shorter probes because of the increased Tms and

enables improved Tm leveling, which increases the specificity of the detection reaction.

Another approach to detection, characterization, and quantification of real-time PCR

amplicons involves the use of a nonstandard DNA base pair constructed from isoG and isoC

(108, 146, 158). These synthetic bases pair with each other, but not with the natural bases

guanine and cytosine, and can be covalently coupled to a wide variety of reporter groups. In

the MultiCode-RTx assays (EraGen Biosciences) the target is amplified using a forward

primer with a single isoC nucleotide with fluorescent label at 5′ end and an unlabeled

standard base reverse primer. Amplification is performed in the presence of isoG coupled to

a fluorescence quencher molecule, and site-specific incorporation by the DNA polymerase

places the quencher in close proximity to the fluorophore, resulting in a decrease of

fluorescence with every PCR cycle. The number of cycles in which the fluorescence change

can be detected is dependent on the initial number of target molecules in the reaction. The

decrease in fluorescence is easily monitored by a number of different standard real-time PCR

instruments. Postreaction amplicon melting-curve analysis can be performed to confirm the

identity of the amplicon and to detect sequence variants. MultiCode-RTx research-use-only

assays for detection of N. gonorrhoeae and for quantification of CMV, Epstein-Barr virus

(EBV), and BK virus (BKV) are available from EraGen.

Real-time PCR methods decrease the time required to perform nucleic acid assays because

there are no post-PCR processing steps. Also, since amplification and detection occur in the

same closed tube, these methods eliminate the postamplification manipulations that can lead

to laboratory contamination with the amplicon. In addition, real-time PCR methods lend

themselves well to quantitative applications because analysis is performed early in the log

phase of product accumulation, and as a result, they are less prone to error resulting from

differences in sample-to-sample amplification efficiency. However, the multiplexing

capabilities of these methods are limited due to the overlapping absorption and emission

spectra of available fluorophores, thus restricting the number of multiplexed targets to four

or five (75).

Digital PCR

PCR exponentially amplifies nucleic acids and the number of amplification cycles, and the

amount of amplicon allows the computation of the starting quantity of targeted nucleic acid.

However, many factors complicate this calculation, often creating uncertainties and

inaccuracies, particularly when the starting concentration is low. Digital PCR attempts to

overcome these difficulties by transforming the exponential data from conventional PCR to

digital signals that simply indicate whether amplification occurred (68, 159, 173).

Digital PCR is accomplished by capturing or isolating each individual nucleic acid molecule

present in a sample within many chambers, zones, or regions that are able to localize and

concentrate the amplification product to detectable levels. After PCR amplification, a count of

the areas containing PCR product is a direct measure of the absolute quantity of nucleic acid

in the sample. The capture or isolation of individual nucleic acid molecules may be done in

capillaries, microemulsions, or arrays of miniaturized chambers or on surfaces that bind

nucleic acids. Digital PCR has many applications, including detection and quantification of low

levels of pathogen sequences, rare genetic sequences, gene expression in single cells, and

clonal amplification of nucleic acids for sequencing mixed nucleic acid samples. Clonal

amplification enabled by digital PCR is a key element of many of the “next-generation”

sequencing methods described below.

Transcription-Based Amplification Methods

Nucleic acid sequence-based amplification (NASBA) and transcription-mediated amplification

(TMA) are both isothermal RNA amplification methods modeled after retroviral replication

(24, 48, 79). The methods are similar in that the RNA target is reverse transcribed into cDNA

and then RNA copies are synthesized with an RNA polymerase. NASBA uses avian

myeloblastosis virus RT, RNase H, and T7 bacteriophage RNA polymerase, whereas TMA uses

an RT enzyme with endogenous RNase H activity and T7 RNA polymerase.

Amplification involves the synthesis of cDNA from the RNA target with a primer containing

the T7 RNA polymerase promoter sequence (Fig. 7). The RNase H then degrades the initial

strand of target RNA in the RNA-cDNA hybrid. The second primer then binds to the cDNA and

is extended by the DNA polymerase activity of the RT, resulting in the formation of dsDNA

containing the T7 RNA polymerase promoter. The RNA polymerase then generates multiple

copies of single-stranded, antisense RNA. These RNA product molecules reenter the cycle,

with subsequent formation of more double-stranded cDNA molecules that can serve as

templates for more RNA synthesis. A 109-fold amplification of the target RNA can be

achieved in less than 2 h by this method.

The single-stranded RNA products of TMA in the Gen-Probe tests are detected by

modification of the hybridization protection assay. Oligonucleotide probes are labeled with

modified acridinium esters with either fast or slow chemiluminescence kinetics so that signals

from two hybridization reactions can be analyzed simultaneously in the same tube. The

NASBA products in the bioMerieux (Durham, NC) tests are detected by hybridization with

probes labeled with tris(2, 29-bispyridine)ruthenium and electrochemiluminescence. NASBA

has also been used with molecular beacons to create a homogeneous, kinetic amplification

system similar to real-time PCR (86).

Transcription-based amplification systems have several strengths, including no requirement

for a thermal cycler, rapid kinetics, and a single-stranded RNA product that does not require

denaturation prior to detection. Also, single-tube clinical assays and a labile RNA product

may help minimize contamination risks. Limitations include the poor performance with DNA

targets and concerns about the stability of complex multienzyme systems. Gen-Probe has

developed TMA-based assays for detection of Mycobacterium tuberculosis, C. trachomatis, N.

gonorrhoeae, HCV, and HIV-1. NASBA-based kits (bioMerieux) for the detection and

quantification of HIV-1 RNA and detection of enterovirus and respiratory syncytial virus RNA

are commercially available. A basic NASBA kit is also available for the development of other

applications defined by the user. In its latest iteration, NucliSens EasyQ, NASBA is coupled

with molecular beacons for real-time amplification and detection of target nucleic acids (12).

Strand Displacement Amplification

Strand displacement amplification (SDA) is an isothermal template amplification technique

that can be used to detect trace amounts of DNA or RNA of a particular sequence. SDA, as it

was first described, was a conceptually straightforward amplification process with some

technical limitations (174). Since its initial description, however, it has evolved into a highly

versatile tool that is technically simple to use but conceptually complex. SDA is the

intellectual property of BD Diagnostics.

In its current iteration, SDA occurs in two discrete phases, target generation and exponential

target amplification (91). Both are illustrated in Fig. 8. In the target generation phase, a

dsDNA target is denatured and hybridized to two different primer pairs, designated as

bumper and amplification primers. The amplification primers include the single-stranded

restriction endonuclease enzyme sequence for BsoB1 located at the 5′ end of the target

binding sequence. The bumper primers are shorter and anneal to the target DNA just

upstream of the region to be amplified. In the presence of BsoB1, an exonuclease-free DNA

polymerase, and a dNTP mixture consisting of dUTP, dATP, dGTP, and thiolated dCTP (Cs),

simultaneous extension products of both the bumper and amplification primers are

generated. This process displaces the amplification primer products, which are available for

hybridization with the opposite-strand products with the opposite-strand bumper and

amplification primers.

The simultaneous extension of opposite-strand primers produces strands complementary to

the product formed by extension of the first amplification primer, with Cs incorporated into

the BsoB1 cleavage site. This product enters the exponential target amplification phase of

the reaction. The BsoB1 enzyme recognizes the double-stranded site, but because one

strand contains Cs, it is nicked rather than cleaved by the enzyme. The DNA polymerase then

binds to the nicked site and begins synthesis of a new strand while simultaneously displacing

the downstream strand. This step re-creates the double-stranded species with the

hemimodified restriction endonuclease recognition sequence, and the iterative nicking and

displacement process repeats. The displaced strands are capable of binding to oppositestrand

primers, which produces exponential amplification of the target sequences.

These single-stranded products also bind to detector probes for real-time detection. The

detector probes are single-stranded DNA molecules with fluorescein and rhodamine labels.

The region between the labels includes a stem-loop structure. The loop contains the

recognition site for the BsoB1 enzyme. The target-specific sequences are located 3′ of the

rhodamine label. In the absence of a specific target, the stem-loop structure is maintained

with the fluorescein and rhodamine labels in close proximity. The net effect is that very little

emission for the fluorescein is detected after excitation. After SDA, the probe is converted to

a double-stranded species, which is cleaved by BsoB1. The cleavage causes physical

separation of the fluorescein and rhodamine labels, which results in an increase in emission

from the fluorescein label.

SDA has a reported sensitivity high enough to detect as few as 10 to 50 copies of a target

molecule (174). By using a primer set designed to amplify a repetitive sequence with 10

copies in the M. tuberculosis genome, the assay is sensitive enough to detect one to five

genome copies from the bacterium. SDA has also been adapted to quantify RNA by adding

an RT step (RT-SDA). In this case, a primer hybridizes to the target RNA and an RT

synthesizes a cDNA molecule. This cDNA can then serve as a template for primer

incorporation and strand displacement. The products of this strand displacement then feed

into the amplification scheme described above. RT-SDA has been used for the determination

of HIV viral load (117). Food and Drug Administration (FDA)-cleared tests using SDA for the

direct detection of C. trachomatis and N. gonorrhoeae in clinical specimens are available

from Becton Dickinson.

The main advantage of SDA is that it is an isothermal process that, unlike PCR, can be

performed at a single temperature after initial target denaturation. This eliminates the need

for expensive thermal cyclers. Furthermore, samples can be subjected to SDA in a single

tube, with amplification times varying from 30 min to 2 h. The main disadvantage of SDA lies

in the fact that, unlike those at which PCR is performed, the relatively low temperature at

which SDA is carried out (52.5°C) can result in nonspecific primer hybridization to sequences

found in complex mixtures such as genomic DNA. Hence, when the target is in low

abundance compared to background DNA, nonspecific amplification products can swamp the

system, decreasing the sensitivity of the technique. However, the use of organic solvents to

increase stringency at low temperatures and the recent introduction of more thermostable

polymerases capable of strand displacement have alleviated much of this problem.

Loop-Mediated Amplification

Loop-mediated amplification (LAMP) is an isothermal method that relies on autocycling

strand displacement DNA synthesis by Bst DNA polymerase and a set of four to six primers

(116). Two inner and two outer primers define the target sequence, and an additional set of

loop primers are added to increase the sensitivity of the reaction. The final products of the

LAMP reaction are DNA molecules with a cauliflower-like structure of multiple loops consisting

of repeats of the target sequence. The products can be analyzed in real time by monitoring

of the turbidity in the reaction tube resulting from production of magnesium pyrophosphate

precipitate during the DNA amplification. Amplification products can also be visualized in

agarose gels after electrophoresis and staining with ethidium bromide or SYBR green.

LAMP has been used successfully in a number of laboratory-developed assays to detect DNA

and RNA viruses (64, 119, 166, 186), differentiate viral subtypes (110, 126), and diagnose

mycobacterial infections (66). Since LAMP is an isothermal process and positive reactions can

be detected by simple turbidity measurements or visualized directly with the naked eye, it

requires no expensive equipment. These attributes make it an attractive technology for

resource-poor s ettings and field use (87). However, primer design for LAMP is more complex

than for PCR, with specialized training and software required for their design. Meridian

Bioscience, Inc. (Cincinnati, OH), has licensed LAMP technology from Eiken Chemical

Company, Ltd., Tokyo, Japan, for the development of infectious-disease diagnostics in the

United States.

Helicase-Dependent Amplification

Helicase-dependent amplification (HDA) is an isothermal process developed by BioHelix,

Beverly, MA, that uses helicase to separate dsDNA and generate single-stranded templates

for primer hybridization and subsequent extension by a DNA polymerase (172). As the

helicase unwinds dsDNA enzymatically, the initial heat denaturation and subsequent

thermocycling steps required by PCR can all be omitted. In HDA, strands of dsDNA are

separated by the DNA helicase and the ssDNA-coated ssDNA-binding proteins. Two

sequence-specific primers hybridize to each border of the target sequence, and a DNA

polymerase extends the primers annealed to the target sequence to produce dsDNA. The two

newly synthesized products are used as substrates by the helicase in the next round of

amplification. Thus, a simultaneous chain reaction proceeds, resulting in exponential

amplification of the selected target sequence.

HDA is compatible with multiple detection technologies, including qualitative and quantitative

fluorescence technologies and with instruments designed for real-time PCR (169).

Furthermore, HDA has shown potential for the development of simple, portable DNA

diagnostic devices to be used in the field or at the point of care (17, 46).

PROBE AMPLIFICATION TECHNIQUES

Probe amplification methods differ from target amplification methods in that the amplification

products contain only a sequence present in the initial probes. Ligase chain reaction (185),

cycling probe technology (37), and cleavase-invader technology (96) are all examples of

probe amplification methods for which diagnostic applications have been developed.

However, diagnostic tests based on ligase chain reaction and cycling probe technology are no

longer available in the United States.

Cleavase-Invader Technology

Invader assays (Hologic, Bedford, MA) are based on a probe amplification method that relies

upon the specific recognition and cleavage of particular DNA structures by cleavase, a

member of the FEN-1 family of DNA polymerases. These polymerases cleave the 5′ singlestranded

flap of a branched base-paired duplex. This enzymatic activity likely plays an

essential role in the elimination of the complex nucleic acid structures that arise during DNA

replication and repair. Since these structures may occur anywhere in a replicating genome,

the enzyme recognizes the molecular structure of the substrate without regard to the

sequence of the nucleic acids making up the DNA complex (88).

In the invader assays, two primers are designed which hybridize to the target sequence in an

overlapping fashion (Fig. 9). Under the proper annealing conditions, the probe

oligonucleotide binds to the target sequence. The invader oligonucleotide is designed such

that it hybridizes upstream of the probe, with a region of overlap between the 3′ end of the

invader and the 5′ end of the probe. Cleavase cleaves the 5′ end of the probe and releases

it. It is in this way that the target sequence acts as a scaffold upon which the proper DNA

structure can form. Since the DNA structure necessary to serve as a cleavase substrate

occurs only in the presence of the target sequence, the generation of cleavage products

indicates the presence of the target. Use of a thermostable cleavase enzyme allows reactions

to be run at temperatures high enough for a primer exchange equilibrium to exist. This

allows multiple cleavase products to form off of a single target molecule.

FRET probes and a second invasive cleavage reaction are used to detect the target-specific

products. Invader technology can be used for genotyping, detection of mutations, and viral

load testing. FDA-cleared assays for detection of pools of high-risk genotypes and types 16

and 18 of HPV in cervical samples are available from Hologic (45).

The invader assay has several inherent advantages. Because the overlap in the invader

probe need be only 1 bp, this technology can easily be adapted to detect point mutations of

interest by designing the overlap region to encompass the mutation to be detected (97). The

detection of these point mutations would not require postreaction restriction digestion, since

the primers would be differentially cleaved on the basis of the presence or the absence of the

mutation in question. This feature could be exploited to track mutations in pathogens

associated with drug resistance or virulence. In addition, unlike amplification techniques such

as PCR, SDA, and TMA, in which the target sequence itself is amplified, the invader assay

does not increase the amount of the target sequence. As a consequence, invader assays are

less prone to problems of false-positive results due to amplicon cross contamination.

POSTAMPLIFICATION DETECTION AND ANALYSIS

Gel Analysis

Visualization of amplification products in agarose gels after electrophoresis and ethidium

bromide staining was the earliest detection method. After gel electrophoresis, DNA is often

transferred onto a nitrocellulose or nylon membrane and hybridized to a specific probe to

increase both the sensitivity and the specificity of detection. Membranes with bound

radiolabeled probes are placed in proximity to X-ray film, and the hybrids are visualized as

dark bands. Enzyme-labeled probes can be visualized through either light or color production

after the addition of the appropriate chemiluminescent or chromogenic substrates. Many of

these nonisotopic approaches are at least as sensitive as isotopic methods and are faster. In

addition, the enzyme-labeled probes are more stable. Although gel electrophoresis and

blotting remain important research tools, these techniques are being replaced by faster and

simpler methods in the clinical laboratory.

Single-strand conformation polymorphism (SSCP) analysis and restriction fragment length

polymorphism (RFLP) analysis have been used to ascertain information about the base

compositions of the amplification products visualized in a gel. In SSCP analysis, the PCR

product is denatured and then subjected to electrophoresis in a nondenaturing gel (122).

Variations in the physical conformations of the PCR products are related to the base

compositions and are detected by differential gel migration. This technique has successfully

been used to detect mutations causing rifampin resistance in M. tuberculosis (161).

RFLP analysis uses restriction endonucleases to cleave amplification products at specific

recognition sites. The fragments are separated by electrophoresis, and the resulting banding

pattern provides information about the nucleic acid sequence. When coupled with a

hybridization reaction, RFLP analysis can also provide information about the location and

number of loci homologous to the probe. Both SSCP analysis and RFLP analysis of short

products have largely been replaced by direct DNA sequencing as this technology has

improved and the costs have decreased.

Capillary Electrophoresis

Capillary electrophoresis allows for accurate size discrimination of fluorescently labeled

nucleic acids from 50 to 1,000 bases with single base precision. PCR and capillary

electrophoresis have been functionally integrated to produce highly multiplexed assays that

can simultaneously detect dozens of targets whose identities are defined by the specific size

of the corresponding amplicons (40).

PrimeraDx (Mansfield, MA) has developed a multiplexed assay for the simultaneous

quantification of CMV, herpes simplex virus (HSV), BKV, human herpesvirus 6 (HHV-6), and

HHV-7 viral loads that integrates PCR and capillary electrophoresis. In this assay,

amplification of the nucleic acid targets is monitored by sampling the PCR during sequential

cycles and separating and quantifying the PCR products by capillary electrophoresis. These

data are used to construct amplification curves. Similar to the case with real-time PCR

amplification, a cycle threshold is determined from the amplification curve for each of the

targets in the exponential phase of amplification. Unlike real-time PCR, where standards in a

separate reaction are used, the PrimeraDx assay uses multiple internal standards in each

reaction to generate calibration curves for each individual assay in the multiplex reaction.

Seegene, Inc. (Seoul, South Korea), has developed a wide variety of multiplexed infectiousdisease

assays, including sexually transmitted disease, HPV genotyping, mycobacterial, and

respiratory pathogen panels (71). The targets of these multiplexed assays are designed to be

discriminated by size and are compatible with several different microfluidic and capillary

electrophoresis systems, including the Agilent (Santa Clara, CA) 2100 Bioanalyzer and

Applied Biosystems (Foster City, CA) sequencers.

Colorimetric Microtiter Plate Systems

Colorimetric microtiter plate (CMP) systems are convenient alternatives to traditional gel and

blotting techniques for detection of amplified products. In these systems, the amplified

product is captured in microtiter plate wells by specific oligonucleotide probes coating the

plastic surface. Bound product is detected by a color change that takes place after addition of

an enzyme conjugate and the appropriate substrate. These systems resemble enzyme

immunoassays and use microtiter plate washers and readers commonly found in clinical

laboratories. CMP systems are more practical and faster than the traditional membrane

hybridization techniques described above.

Several variations of CMP systems are commercially available. In one popular approach,

biotinylated primers are used to amplify the target, and the biotin-containing PCR product is

denatured and added to the microtiter well. After hybridization with a capture probe, the

bound product is detected with a streptavidin-enzyme conjugate and a chromogenic

substrate (94). Enzyme-conjugated antibodies directed against dsDNA have also been used

to detect PCR products in CMP systems (100). Another approach uses digoxigenin-dUTP to

label the PCR product and enzyme-conjugated antidigoxigenin antibodies to detect the

captured product (131).

Allele-Specific Hybridization

Line probe assays are manufactured by Innogenetics (Ghent, Belgium) for genotyping of

HCV, HPV, and HBV; identification of mycobacteria; and analysis for drug resistance

mutations in HIV-1, HBV, M. tuberculosis, andHelicobacter pylori (138, 156, 157). The HCV

line probe assays are distributed by Siemens. In these assays, a series of probes with

poly(T) tails are attached to nitrocellulose strips. Biotin-labeled PCR product is then

hybridized to the immobilized probes on the strip. The labeled PCR product hybridizes only to

the probes that give a perfect sequence match under the stringent hybridization conditions

used. After hybridization, streptavidin labeled with alkaline phosphatase is added and binds

to the biotinylated hybrids. Incubation with a chromogen results in a purple precipitate. The

pattern of hybridization provides information about the nucleic acid sequence of the

amplicon. This method is capable of detecting single-nucleotide polymorphisms.

A line probe for identification of 37 HPV genotypes is available from Roche (152). The

method employs multiplex PCR with biotinylated primers targeted to the L1 region of the

HPV genome and a linear array of L1 sequence-specific probes fixed to a nitrocellulose strip.

The pattern of hybridization provides the genotype and is determined as described above.

Direct Sequencing

The combination of PCR and Sanger dideoxynucleotide chain termination methods can be

used to determine DNA sequences in clinical samples (65). The use of fluorescent dye

terminator chemistry and laser scanning in a polyacrylamide gel electrophoresis format has

been the standard in electrophoretic separation technology. However, the recent application

of capillary electrophoresis techniques to the separation of PCR and dideoxy chain

termination products has streamlined the sequencing process by eliminating some of the

labor-intensive steps, which makes the technology a better fit for diagnostic applications

(36). The Clinical and Laboratory Standards Institute (CLSI) has developed guidelines for

nucleic acid sequencing in clinical laboratories (21,112).

CLIP, a coupled amplification and sequencing method, uses oligonucleotide primers labeled

with different fluorescent dyes, standard dideoxynucleotide termination reagents, and PCR to

produce extension products that end with a chain-terminating nucleotide. The nucleic acid

sequence is deduced from the electrophoretic mobilities of the different extension products

from a set of four reactions, each product containing a different chain-terminating

nucleotide. A unique feature of CLIP sequencing is that one reaction produces sequence

information for both nucleic acid strands. CLIP sequencing serves as the basis for

commercially available assays for HIV-1 drug resistance (Siemens Healthcare Diagnostics).

The ViroSeq HIV-1 genotyping assays (Celera Diagnostics, Alameda, CA; distributed by

Abbott Molecular Diagnostics, Des Plaines, IL) also use dideoxy chain-terminating

sequencing, but each dideoxynucleotide is labeled with a different fluorescent dye. Each

reaction mixture contains one primer but all four uniquely labeled dideoxynucleotides.

Separation of the terminated PCR products is done by capillary electrophoresis.

Although direct sequencing of PCR products by electrophoresis is a powerful research tool, its

routine use in the clinical laboratory depends upon the development of high-throughput

systems with integrated databases and data analysis software. Such systems are available

for HIV-1 and HCV genotyping and for identification of bacteria and fungi by rRNA gene

sequence analysis.

Pyrosequencing (Qiagen, Hilden, Germany) represents an alternative approach to

conventional sequencing and is useful for genotyping and short-read-length sequencing (29).

Pyrosequencing is based on the luminometric detection of pyrophosphate that is generated

during DNA synthesis.

A sequencing primer is hybridized to a single-stranded PCR amplicon and incubated with the

enzymes DNA polymerase, ATP sulfurylase, luciferase, and apyrase and the substrates

adenosine 5′-phosphosulfate and luciferin. The first of four dNTPs is added to the reaction

mixture. DNA polymerase catalyzes the incorporation of the dNTP into the DNA strand. Each

incorporation event is accompanied by release of pyrophosphate (PPi) in a quantity equimolar

to the amount of incorporated nucleotide. The ATP sulfurylase quantitatively converts PPi to

ATP in the presence of adenosine 5′-phosphosulfate. This ATP drives the luciferase-mediated

conversion of luciferin to oxyluciferin, which generates light in amounts that are proportional

to the amount of ATP. The light produced in the reaction is detected by a charge-coupled

device camera. A program is produced in which the height of each peak is proportional to the

number of nucleotides incorporated. Apyrase, a nucleotide-degrading enzyme, continuously

degrades ATP and unincorporated dNTPs. This degradation switches off the light and

regenerates the reaction solution. The next dNTP is added, and the process is repeated.

Pyrosequencing has been used in microbiology to detect drug resistance mutations and to

identify and type bacteria, viruses, and fungi (2, 44, 49, 121). Unlike conventional

sequencing strategies, pyrosequencing provides reliable data for sequences adjacent to the

sequencing primer termini. Pyrosequencing provides a simple-to-use and robust platform for

short-read-length sequencing.

Multiple new sequencing technology platforms have emerged since 2005 and have greatly

surpassed conventional dideoxynucleotide chain termination methods in terms of increased

total sequence production and decreased cost. Collectively, these new sequencing methods

are referred to as next-generation sequencing, and they have considerable potential for

clinical diagnostics (163). The three major next-generation sequencing platforms as of this

writing are the Roche 454 GS-FLX (454, Branford, CT), the Illumina (San Diego, CA) Genome

Analyzer, and the ABI SOLiD (Applied Biosystems). The sequencing methods, read lengths,

run times, and total bases per run for each of these methods are compared with those of

Sanger sequencing in Table 1. These new approaches to sequencing are all based on cyclic

stepwise sequencing of massive numbers of templates in parallel in a flow cell and share

similar sample preparation steps, including target DNA fragmentation, ligation to adaptor

sequences, and clonal amplification of targets.

The Roche 454 platform, using pyrosequencing technology described earlier to carry out

hundreds of thousands of sequencing reactions simultaneously on independent beads, works

as follows. Target DNA is first randomly sheared into fragments and then ligated to adaptors.

Single-stranded template DNA is isolated, mixed with beads, and then subjected to emulsion

PCR to clonally amplify the template on each bead. The beads are then distributed into a

“picotiter plate” that contains millions of tiny wells. Within each well that receives an

individual bead, an isolated environment is created for the sequencing of each template,

resulting in massively parallel sequencing of different templates simultaneously.

DNA templates sequenced with the Illumina Genome Analyzer are ligated to adaptor

sequences which serve as priming sites for PCR and sequencing, and to attach

complementary templates to a solid surface through a mechanism called bridge

amplification. Bridge amplification generates clusters of amplified template on the solid

surface, where each cluster represents a different template. It uses a unique sequencing

chemistry that incorporates fluorescently labeled, reversible terminator nucleotides. These

nucleotides are labeled with different color fluorophores so that all four nucleotides can be

added to the reactions simultaneously. Only one terminator nucleotide can be incorporated

into each sequence during one sequencing cycle, and the color of the fluorescent label

incorporated into the sequences of each cluster is recorded. Removal of the terminator group

on the nucleotide just added enables incorporation of the next complementary nucleotide,

and the cycle is repeated.

The ABI SOLiD System chemistry starts with emulsion PCR of adaptor-modified ssDNA

molecules. After PCR, the templates are denatured and bead enrichment is performed to

select beads with extended templates. The template on the selected beads undergoes a 3′

modification to allow covalent binding to a glass slide. The modified beads are deposited

randomly on the slide. The sequencing occurs by ligation. Primers hybridize to the adaptor

sequence within the library template. A set of four fluorescently labeled di-base probes

compete for ligation to the sequencing primer. Specificity of the di-base probe is achieved by

interrogating every first and second base in each ligation reaction. Multiple cycles of ligation,

detection, and cleavage are performed, with the number of cycles determining the eventual

read length. Following a series of cycles, the extension product is removed and the template

is reset with a primer complementary to the n- 1 position for a second round of ligation

cycles. Five rounds of primer resets are completed for each sequencing tag. Consequently,

each base is interrogated in two independent ligation reactions by two different primers. This

dual interrogation provides highly accurate sequences.

Next-generation sequencing will have a major impact on genomics research. In the field of

medical microbiology, applications are evolving in the areas of metagenomics, microbial

identification, and detection of rare mutations. Ultradeep sequencing using the Roche 454

system can detect rare viral variants consisting of as little as 1% of the population, levels far

deeper than those achievable with traditional sequencing methods, and the detection of

these low-abundance drug resistance mutations may significantly impact treatment

outcomes in HIV-1 infections (147, 177).

Hybridization Arrays

High-Density Arrays

High-density DNA hybridization arrays are produced by attaching or synthesizing hundreds or

thousands of oligonucleotides on a solid support in precise patterns. A labeled amplification

product is hybridized to the probes, and hybridization signals are mapped to various

positions within the array. If the number of probes is sufficiently large, the sequence of the

PCR product can be deduced from the pattern of hybridization (resequencing arrays). A

number of manufacturers have developed high-density DNA microarrays and the

instrumentation required to acquire and analyze the data. Hybridization arrays have a

number of applications in microbiology, including microbial and host gene expression

profiling and diagnostic sequencing. The CLSI has published a guideline for the use of

diagnostic nucleic acid microarrays (20).

One of the most developed approaches brings together advances in synthetic nucleic acid

chemistry with photolithography, a process used in the manufacture of semiconductors for

the computer industry (Affymetrix, Santa Clara, CA). This approach uses light to direct the

synthesis of short oligonucleotides on a silica wafer (127). On a 15-mm-square chip,

thousands of individual sites or features can be established. At each feature, specific

oligonucleotides are assembled one nucleotide at a time by light-activated chemistry.

There are a variety of sample preparation methods for the different array types, but all share

a few fundamental characteristics. All methods start with extraction of total RNA, poly(A), or

genomic DNA that is then converted to either cDNA or cRNA by enzymatic methods that

modestly amplify the sample with tagging or incorporating biotinylated or fluoresceinated

nucleotides. In expression applications, the amplification must maintain the relative

abundance levels of the different transcripts present, whereas for re sequencing applications,

the relative abundance of information is rarely important. The DNA chip is hybridized in a

flow cell with the sample for 2 to 12 h. After hybridization, a scanning laser confocal

microscope evaluates the surface fluorescence intensity of the chip. Automated scanning by

the microscope takes only a few minutes to acquire an image of the entire surface of the

chip, and computer software analyzes the fluorescent image and determines the nucleic acid

sequence or gene expression profile of the sample.

Another method of producing DNA hybridization arrays involves the precise micropipetting of

premade dsDNA probes (typically 200 to 2,000 bp in length) onto glass slides with a robotic

device (144). These arrays are not suitable for mutation detection due to the size and

density of the arrayed DNA probes but have facilitated gene expression profiling. DNA arrays

of this type can be used to determine the activation states (mRNA levels) of thousands of

genes simultaneously. Gene expression profiling of pathogens by use of arrays may provide

new insights into pathogenic mechanisms and help identify new therapeutic and vaccine

targets.

High-density microarrays coupled with sequence-independent PCR have also been used in

the discovery and characterization of pathogens and have the potential to provide rapid,

unbiased, differential diagnoses of infectious diseases. Wang et al. described the first

microarray designed to detect large numbers of viruses (178). The microarray consisted of

1,600 70-mer oligonucleotides derived from 140 different virus species, with an average of

10 oligonucleotides per virus species. They demonstrated that a wide variety of viruses could

be detected by the microarray with sensitivities and specificities similar to those of individual

virus-specific PCR assays (14). In addition, this approach has facilitated the discovery of a

number of novel viruses from humans and animals, including the severe acute respiratory

syndrome coronavirus (179). The field of diagnostic micorarrays is rapidly developing, with

multiple broad-range microarrays described (73, 90, 123, 183).

High-density microarrays hold much promise for molecular diagnostics. However, the

complexity of fabricating the arrays, limited availability, and high test costs are obstacles to

their routine use in clinical laboratories.

Low- to Moderate-Density Arrays

Recent developments of new detection techniques and simplified methodologies have

facilitated the transition from expensive high-density arrays to cost-effective, low- to

medium-density systems for clinical diagnostics. The three microarray systems described in

the following paragraphs all are FDA-cleared platforms for human genetic and

pharmacogenetic applications. Each of the manufacturers has infectious-disease applications

under development.

The INFINITI analyzer (Autogenomics, Carlsbad, CA) is a fully automated, multiplexing

platform that uses novel BioFilmChip microarrays for a wide range of molecular diagnostic

applications. Fluorescence-labeled PCR amplicons are hybridized to probes immobilized on a

BioFilmChip microarray. The microarray is film-based microarray, which consists of multiple

layers of thin hydrogel matrices on a polyester solid support. Each spot on the array is

scanned with a built-in confocal microscope. The system has integrated controls for all steps

and automatically processes and analyzes data. Infectious-disease applications under

development include microarrays for detection of drug resistance in M. tuberculosis,

respiratory viruses, sexually transmitted disease agents, and nontuberculous mycobacteria.

The Verigene system (Nanosphere, Inc., Northbrook, IL) uses gold nanoparticle-labeled

probes to detect target nucleic acid hybridized to capture oligonucleotides arrayed on a glass

slide. Silver signal amplification is then performed on the gold nanoparticle probes that are

hybridized to the captured DNA targets of interest. The Verigene Reader optically scans the

slide for silver signal, processes the data, and produces a qualitative result. A respiratory

virus panel for the Verigene system that detects influenza A and B viruses and respiratory

syncytial virus has been cleared by the FDA.

The eSensor system (Osmetech Molecular Diagnostics, Pasadena, CA) uses electrochemicaldetection-

based DNA microarrays (93). These microarrays are composed of a printed circuit

board consisting of an array of 76 gold-plated electrodes. Each electrode is modified with a

multicomponent, self-assembled monolayer that includes presynthesized oligonucleotide

capture probes. Nucleic acid detection is based on a sandwich assay principle. Signal and

capture probes are designed with sequences complementary to immediately adjacent regions

on the corresponding target DNA sequence. A three-member complex is formed between

capture probe, target sequence, and signal probe based on sequence-specific hybridization.

This process brings the 5′ end of the signal probe containing electrochemically active

ferrocene labels into close proximity to the electrode surface.

The ferrous ion in each ferrocene group undergoes cyclic oxidation and reduction, leading to

loss or gain of an electron, which is measured as current at the electrode surface using

alternating-current voltammetry. Higher-order harmonic signal analysis also facilitates

discrimination of ferrocene- dependent faradic current from background capacitive current.

The eSensor cartridge consists of a printed circuit board, a cover, and a microfluidic

component. The microfluidic component includes a diaphragm pump and check valves in line

with a serpentine channel that forms the hybridization channel above the array of electrodes.

The eSensor instrument consists of a base module and up to three cartridge-processing

towers, each with eight slots for cartridges. The cartridge slots operate independently of

each other. The throughput of a three-tower system can reach 300 tests in 8 h. A respiratory

pathogen panel for the eSensor system is currently under development.

Mass Spectrometry

One of the most exciting developments in molecular microbiology is the application of mass

spectrometry to identification and characterization of pathogens. Mass spectrometry is

remarkably sensitive and accurate, with a throughput exceeding one sample per minute.

Mass spectrometers are now common in clinical laboratories, and the advent of smaller,

lower-cost instruments could facilitate wider use. Fully integrated systems for infectiousdisease

applications are available from Ibis Biosciences (Carlsbad, CA), a subsidiary of

Abbott Molecular, and Sequenom (San Diego, CA).

Mass tag PCR uses a library of 64 distinct MassCode tags (Qiagen) to code different gene

targets in multiplex PCRs. Target nucleic acids are amplified by multiplex PCR using primers

labeled by a photocleavable link to molecular tags of different molecular weights. After

removing the unincorporated primers, tags are released by UV irradiation and analyzed in a

single quadropole mass spectrometer. The identity of the gene target in the clinical samples

is determined by the size of its cognate tags. This approach was used to develop a rapid,

sensitive, multiplex assay for the detection of 22 different respiratory pathogens in clinical

samples (9).

The T5000 Universal Biosensor (Ibis) is a commercially available system capable of

identification and characterization of a broad range of pathogens (31). In this system all

nucleic acids present in a clinical sample are extracted and aliquoted into wells of a microtiter

plate that each contain one or more pairs of broad-range primers for PCR. The primers are

designed to amplify a product from a selected group of microorganisms, for example, all

bacteria, specific species, or individual strains. The PCRs produce a mixture of products

reflecting the complexity of the original mixture of microorganisms present in the clinical

sample.

The PCR products are desalted and sequentially electrosprayed into a mass spectrometer for

analysis. The spectral signals are processed to determine the masses of each of the PCR

products present with sufficient accuracy that the base composition of each amplicon can be

unambiguously deduced. Using the combined base compositions from multiple PCRs, the

identities of the pathogens and their relative concentrations in the sample can be

determined. Although it is not immediately intuitive, nucleic acid composition (i.e., the

numbers of a ’s, G’s, C’s, and T’s) in specific regions of the genome is equally as informative

as the nucleic acid sequence. Mass spectrometry is remarkably sensitive and can measure

the weight and determine the base composition from small quantities of nucleic acids in

complex mixtures essentially instantaneously. A key element of the Ibis system is a curated

database of genomics that associates base counts with primer pairs for thousands of

organisms. Broad-range PCRs are capable of producing products from groups of organisms

rather than single species. That, coupled with the ability of the mass spectrometer to rapidly

and accurately derive base compositions from PCR amplicons, provides high information

content and eliminates the need to anticipate which pathogen is present in the sample. The

Ibis system has been used for the rapid identification and strain typing of a variety of

bacteria, viruses, fungi, and protozoa (32, 141, 142).

Sequenom developed comparative sequencing by base-specific cleavage and matrix-assisted

laser desorption ionization–time-of-flight (MALDI-TOF) mass spectrometry for automated,

high-throughput microbial DNA sequence analysis (62). In this innovative genotyping

method, PCR-amplified signature sequences are subjected to in vitro transcription and basespecific

RNA cleavage by RNase A. Mass signal patterns of the resulting cleavage products, a

mixture of RNA fragments known as compomers, are acquired and provide a fingerprint of

the microorganism. Each RNA compomer is defined by its nucleotide composition with the

cleavage base terminating its 3′ end and thus by its mass in the resulting mass spectrum.

The list of detected experimental compomer masses is compared with a calculated list of

molecular weights derived from an in silico digest of a set of reference sequences in the

system database. The simulated patterns of the reference set are used to identify the

microorganism by its best match to a reference sequence. Small differences between the

best-matching reference and sample sequences show up as a difference between the in silico

and detected sample spectra. They can be used to identify and localize sequence differences

down to a single base change and identify novel sequences. Depending on the gene target,

MALDI-TOF mass spectrometry can provide high-level discrimination of individual microbial

taxa or be used to identify lineages within a species (84, 92, 149, 155).

QUANTITATIVE METHODS

Many of the methods discussed above can be used to quantify the amount of RNA or DNA in

a clinical sample. The most commonly used methods include PCR and RT-PCR, transcriptionbased

amplification, and bDNA assays. The principle of quantitative molecular methods is

that there is a linear relationship between the quantity of the input template and the amount

of the product or signal generated. Competitive PCR (cPCR) is a reliable and robust method

that was the basis of the first generation of viral load assays for HIV-1 and HCV (Roche

Amplicor Monitor System) that were commonly used in clinical laboratories. These assays,

based on conventional standard PCR, are still in use by clinical laboratories but are rapidly

being replaced by real-time amplification methods. The basic concept behind cPCR is the

coamplification in the same reaction tube of target and calibrator templates with equal or

similar lengths and with the same primer binding sequences (18). Since the templates are

amplified with the same primer pair, identical thermodynamics and amplification efficiencies

are ensured. The amount of the calibrator must be known, and after amplification, products

from the templates must be distinguishable from each other. Different types of calibrators

have been used in cPCR, but in general those calibrators similar in size and base composition

to the target work most effectively. RNA competitors should be used in quantitative RT-PCRs

to address the problem of variable RT efficiency. This competitive amplification approach has

also been used effectively with transcription-based amplification methods using RNA targets

and RNA calibrators.

For cPCR, the concentration of the target template in the clinical sample can be determined

by a simple calculation. The yield of the PCR product is described by the equation Y = I(1

+ e)n, where Y is the quantity of the PCR product, I is the quantity of the template at the

beginning of the reaction, e is the efficiency of the reaction, and n is the number of cycles. In

cPCR, this equation is written for both templates, as follows: competitor, Yc = Ic (1 + e)n;

target, Yt = It (1 + e)n. Since e and n are the same for both the competitor and the target,

the relative product ratio, Yc/Yt, directly depends on the initial concentration ratio, Ic/It, and

the function, Yc/Yt = Ic/It, is linear.

Real-time amplification and detection methods are particularly well suited for quantification

of nucleic acid because the amount of the fluorescent signal generated is proportional to the

concentration of the target DNA or RNA in the original sample. Real-time PCR and

transcription-based amplification methods are the most commonly used quantitative

methods. For real-time PCR, the fluorescent signal is measured during the exponential phase

of amplification, which is where the amplification plot crosses the threshold (Fig. 3). This is in

contrast to standard PCR methods that measure the end point signal. There are advantages

to measuring the fluorescent signal during the exponential phase of amplification; the

reaction components are not limiting, and the assay is less sensitive to the effects of

inhibitors. As a result, real-time PCR assays are more reproducible than standard PCR

assays. Both internal and external calibrators can be used with real-time assays, but the

improved precision of real-time assays allows more reliable results to be obtained with an

external calibration curve than would be obtained with standard PCR. When external

calibrators are used, a calibration curve is generated by plotting the log10 concentration of

the external calibrator versus the CT, and this plot is used to calculate the concentration of

nucleic acid in the sample. The concentration of nucleic acid in the sample is inversely

related to the CT: the higher the concentration of the nucleic acid, the lower the CT(59). In

general, quantitative real-time PCR assays are not more sensitive than standard PCR assays;

however, they have a much broader linear range, typically 6 to 7 orders of magnitude.

The CLSI has published guidelines for quantitative molecular methods for infectious diseases

that address the development and application of quantitative PCR assays and other nucleic

acid amplification methods (111).

AUTOMATION AND INSTRUMENTATION

Molecular assays consist of three major steps: specimen processing, nucleic acid

amplification, and product detection. Sample processing is usually the most labor-intensive

step and has represented the biggest challenge for manufacturers of automated test

systems. However, in the past several years there have been considerable advances in this

area with the availability of both semiautomated and fully automated systems. Automation of

the nucleic acid extraction process offers laboratories several advantages, including ease of

use, limited handling of the sample, improved reproducibility, reduced opportunity for cross

contamination, and, for some systems, postelution functions such as adding samples into the

master mix. These advantages need to be weighed against the costs of automated systems,

the inflexibility of batch size, and the large sizes of many of the automated instruments. The

systems vary in the types of nucleic acid extraction methods that they provide and include

total nucleic acid, DNA-only, and RNA-only protocols. Other features of automated extraction

systems to consider are the availability of protocols for various specimen types and volumes,

variable elution volumes, the availability of target-specific and/or g eneric target extraction

methods, and specimen throughput. The available automated systems range from fully

automated high-throughput systems such as the MagNA Pure system (Roche) and m2000

generic extractor (Abbott) to those designed for a small number of specimens with randomaccess

capabilities, such as BioRobot EZ1 (Qiagen).

There are a few automated systems available for the conventional amplification methods,

such the COBAS system (Roche) for PCR and the System 340 and 440 platforms for bDNA

assays (Siemens). Considerable advances in automation have been made with the

availability of real-time amplification and detection systems.

Several instruments are commercially available for real-time PCR testing. These instruments

vary as to speed, capacity of samples per test run, reaction volume, optics, and support for

different fluorescence probe types. The time required for analysis depends to a great extent

on the time required for thermocycling, and the speed of thermocycling depends on how

quickly the instrument can change temperature over time. For example, some instruments

can change the temperature at 20°C per s, permitting instrument analysis of up to 32

samples in as little as 30 min. Capacity may offset thermocycling speed. Although a highercapacity

instrument may have a longer thermocycling time than a lower-capacity instrument,

potentially more samples may be analyzed by the high-capacity instrument in a specific time

period than by the low-capacity instrument.

The reaction mixture volume assayed may also vary from one system to another. If target

nucleic acid is present in extremely small amounts in a sample, an instrument that permits

higher-volume analysis may be preferred.

Real-time PCR instruments utilize a variety of optics for fluorescence detection. A tungsten

source lamp for excitation and selectable filters for excitation and emission wavelength

detection are used in a number of instruments. Light-emitting diodes or laser excitation

devices coupled with emission wavelength detection may also be used. The new real-time

PCR instruments allow up to six different fluorogenic dyes to be used simultaneously in one

reaction. Until recently, real-time PCR instruments were designed for research applications.

The Prism series of sequence detection systems (Applied Biosystems), LightCycler (Roche),

SmartCycler (Cepheid, Sunnyvale, CA), and Rotor-Gene (Qiagen) are examples of research

instruments that find widespread use in molecular diagnostics laboratories. The COBAS

TaqMan analyzer (Roche) and them2000 system (Abbott) are the first real-time instruments

designed specifically for use in clinical laboratories (4).

Many manufacturers are coupling automated nucleic acid extraction instruments with

amplification and detection systems to create high-throughput, fully automated nucleic acid

analyzers. The TIGRIS system (Gen-Probe), the AmpliPrep-COBAS TaqMan system (Roche),

the m2000 system (Abbott), and the Viper System (BD Diagnostics) are examples of fully

automated and integrated systems designed to perform sample processing, nucleic acid

amplification, and product detection. The GeneXpert system (Cepheid) represents the other

end of the automation spectrum, in which a single sample is added to a disposable fluidic

cartridge that fully automates and integrates sample preparation, amplification, and realtime

detection. The instrument is a random-access design, amendable to on-demand

molecular diagnostic testing.

CURRENT APPLICATIONS

Molecular methods have created new opportunities for the clinical microbiology laboratory to

affect patient care in the areas of initial diagnosis, disease prognosis, and m onitoring of

response to therapy. Over time the methods have become more automated, the cost of

testing has decreased, and clinical utility has been proven for the diagnosis and management

of a variety of infectious diseases. As a result, molecular testing is now routinely performed

in many clinical microbiology laboratories, and clinical applications will continue to expand in

the future.

Initial Diagnosis

With the development of molecular methods, the clinical microbiology laboratory is no longer

reliant solely on the traditional culture methods for detection of pathogens in clinical

specimens. Culture-based methods have long been the gold standard for infectious-disease

diagnosis, but for several diseases, nucleic acid-based tests have replaced culture as the gold

standard. HCV infection, enteroviral meningitis, pertussis, HSV encephalitis, and genital

infections due to C. trachomatis are some examples of infectious diseases in which nucleic

acid-based tests are the new gold standards for diagnosis. This technology has been used to

best advantage in situations in which traditional methods are slow, insensitive, expensive, or

not available. These techniques work particularly well with fragile or fastidious

microorganisms that may die in transit or be overgrown by contaminating biota when

cultured. N. gonorrhoeae is an example for which the nucleic acid can be detected under

circumstances in which the organism cannot be cultured. The use of improper collection

media, inappropriate transport conditions, or delays in transport can reduce the viability of

the pathogen but may leave the nucleic acid still detectable. It is beyond the scope of this

chapter to review all of the possible applications or to provide a compendium of methods for

detection of various pathogens. The reader is directed to another excellent resource for this

information (129).

Opportunities to actually replace culture for bacterial pathogens in routine practice are

limited by the need to isolate the organisms for antibiotic susceptibility testing. In those

applications in which culture has actually been replaced by nucleic acid testing, the

pathogens are of predictable susceptibilities and consequently, routine susceptibility testing

is not performed, or the genetics of resistance are well defined and simple to detect, such as

methicillin resistance in S. aureus.

Molecular methods have had the biggest impact in clinical virology, in which the molecular

approaches are often faster, more sensitive, and more cost-effective than the traditional

methods. The diagnoses of enteroviral meningitis, HSV encephalitis, and CMV infections in

immunocompromised patients are examples of clinically relevant and cost-effective

applications of nucleic acid-based tests. There are greater opportunities to replace the

conventional methods in virology than in bacteriology because the culture-based methods

are costly and antiviral susceptibility testing is not routinely performed. In those situations in

which antiviral susceptibility testing is required, such as identification of ganciclovir-resistant

CMV, molecular methods (i.e., sequencing) are the method of choice for rapid identification

of mutations. The diagnostic role of molecular tests has been further expanded with the FDA

approval of the APTIMA HIV-1 RNA qualitative test (Gen-Probe). The diagnosis of HIV-1

infection has traditionally been performed with screening serologic tests and Western blotting

as the confirmatory test. The APTIMA test can now be used for the diagnosis of acute HIV-1,

to resolve indeterminate Western blot results, and to confirm the screening serologic result.

A limitation of molecular tests for viral diagnostics is the clinical need for simultaneous

identification of multiple pathogens, for example, respiratory viruses. Recently, multiplex

PCR tests have been developed and some FDA cleared that allow for the detection of several

respiratory viruses in a single test. Real-time PCR tests utilize multiple primer pairs and

probes with different fluorescent dyes to detect influenza A and B viruses and respiratory

syncytial virus, as well as an internal control (85). Using this approach, multiple tests will be

needed to detect the common respiratory viruses that are now identified using fluorescentantibody

testing and culture. A second approach is to perform a conventional multiplex PCR

utilizing primer pairs targeting a larger number of viruses and coupling this with the Luminex

bead detection system described above (98, 113). This allows for the detection of many

respiratory viruses in a single test but requires postamplification manipulation of the sample,

which introduces the possibility of false-positive results due to carryover contamination. It is

likely that both of these multiplex approaches will be applied for other groups of pathogens,

such as those causing central nervous system infections and diarrheal diseases.

Perhaps the greatest impact of molecular methods has been in the discovery of previously

unrecognized or uncultivable pathogens. During the past 20 years, a number of infectious

agents were first identified directly from clinical material by using molecular methods. HCV,

the principal etiologic agent of what was once known as non-A, non-B hepatitis, was

discovered in 1989 through the application of molecular cloning techniques by investigators

from the Centers for Disease Control and the Chiron Corporation (16). Cloning and analysis

of the HCV genome led to production of viral antigens that now serve as the basis of the

specific serologic tests used to screen the blood supply and to diagnose hepatitis C. To date,

HCV has resisted all attempts at sustained in vitro propagation. As a result, RT-PCR is used

to detect, quantify, and genotype HCV in infected individuals.

Tropheryma whipplei, the causative agent of Whipple’s disease, is another example of an

uncultivable microorganism which was initially identified by molecular methods (134). It was

discovered by the use of broad-range PCR, in which primers are directed against conserved

sequences in the bacterial 16S rRNA gene. Sequence analysis of the PCR product and

comparison with known 16S rRNA gene sequences were used to characterize the organism

and establish its disease association. This approach provides a new paradigm for discovery of

unrecognized pathogens that is of value in other diseases with features that suggest an

infectious etiology.

Molecular methods are very powerful tools for the identification of emerging pathogens. RTPCR

with consensus primers was used to rapidly identify the etiologic agent of severe acute

respiratory syndrome as a coronavirus (76, 128). Within a few months of the recognized

outbreak, the virus was identified and sequenced and the molecular assays were developed

that played an essential role in diagnosing the infection and defining the epidemiology of the

infection. Similarly, high-throughput sequencing has been used to identify a novel

polyomavirus, WU virus, from a nasopharyngeal aspirate from a 3-year-old with pneumonia

(42). Using a specifically designed real-time PCR assay, this virus has been shown to be

present in 0.7 to 3.0% of patients with acute respiratory infections; the majority of patients

were coinfected with other respiratory viruses (42, 82).

Identification of Bacteria and Fungi by Nucleic Acid

Sequencing

Nucleotide sequence analysis of the 16S bacterial rRNA gene has expanded our knowledge of

the phylogenetic relationships among bacteria and is the new standard for bacterial

identification. rRNA contains several functionally different regions, with some regions having

highly conserved and others having highly varied nucleic acid sequences (181). The

sequence of the 16S rRNA gene is a stable genotypic signature that can be used to identify

an organism at the genus or species level. The 16S gene sequence can be determined

rapidly and provides objective results independent of phenotypic characteristics. As

discussed in the preceding section, it can also be used to characterize previously

unrecognized species. A similar approach that targets the nuclear large subunit of the rRNA

gene can be used for the identification of fungi (77). This gene is found in all fungi and

contains sufficient variation to identify most fungi accurately to the species level.

The DNA sequencing approach to microbial identification involves extraction of the nucleic

acids, amplification of the target sequence by PCR, sequence determination, and a computer

software-aided search of an appropriate sequence database. The major limitations of this

approach to microbial identification include the high cost of automated nucleic acid

sequencers, the lack of appropriate analysis software, and limited databases.

Applied Biosystems has developed ribosomal gene sequencing kits for bacteria and fungi. A

sequence from an unknown bacterium is compared with either full or partial 16S rRNA

sequences from over 1,000 type strains by using the MicroSeq analysis software (160). The

software analysis provides percent base pair differences between the unknown bacterium

and the 20 most closely related bacteria, alignment tools to show differences between the

related sequences, and phylogenetic tree tools to verify that the unknown bacterium actually

clusters with the 20 closest bacteria in the database. The MicroSeq fungal identification

system is similar to the bacterial identification system but targets D2 large-subunit rRNA

(52, 53). Continued improvements in automation, refinements of analysis software, and

decreases in cost should lead to more widespread use of nucleic acid sequence-based

approaches to microbial identification.

More recently, pyrosequencing, or sequencing by synthesis, has been used for the

identification of infectious pathogens. Since the length of high-quality sequence generated is

limited to 50 to 100 bp, it is very useful for single-nucleotide polymorphism analysis, but it

has also been applied to taxonomic categorization of microorganisms. This approach requires

identifying a variable region that contains a unique sequence for the different

microorganisms within the group. Pyrosequencing has been successfully used to classify

mycobacteria and nocardiae into clinically important groups and to identify yeasts and

filamentous fungi (132,170).

Disease Prognosis

Molecular techniques have created opportunities for the laboratory to provide important

information that may predict disease progression. Probably the best example is HIV-1 viral

load as a predictor of progression to AIDS and death in infected individuals. This predictive

value was first demonstrated in 1996 as part of a multicenter AIDS cohort study (103). The

investigators showed that the risk of progression to AIDS and death was directly related to

the magnitude of the viral load in plasma at study entry. The viral load in plasma was a

better predictor of disease progression than the number of CD4+ lymphocytes. Subsequent

studies have confirmed that baseline viral load critically influences disease progression.

Subtyping of certain viruses by molecular methods may also have prognostic value.

Subtyping of respiratory syncytial viruses may provide information about the severity of

infection in hospitalized infants, with those infected with group A viruses having poorer

outcomes (176). HPV causes dysplasia, intraepithelial neoplasia, and carcinoma of the cervix

in women. HPV types 16 and 18 are associated with a high risk of progression to neoplasia,

and types 6 and 11 are associated with a low risk of progression (133). The clinical utility of

molecular testing for high-risk HPV DNA has been established for managing women with the

cervical cytologic diagnosis of atypical squamous cells of undetermined significance. Women

with this condition can be referred for colposcopy based on the detection of high-risk HPV

DNA (151). HPV DNA testing is approved by the FDA for use as an adjunct to cytology for

cervical cancer screening in women aged 30 years or more (184).

CMV viral load testing has recently been shown to be useful for deciding when to initiate

preemptive therapy in organ transplant recipients and distinguishing active disease from

asymptomatic infection. Studies have shown that the level of CMV DNA can predict the

development of active CMV disease (33, 63), with higher viral load values increasing the risk

of symptomatic disease. It is likely that quantitative assays will be also useful in

distinguishing disease from infection with other herpesviruses such as Epstein-Barr virus and

HHV-6.

Duration of and Response to Therapy

Molecular methods have been developed to detect the genes responsible for resistance to

single antibiotics or classes of antibiotics in bacteria and in many cases are superior to the

phenotypic, growth-based methods. The detection of meth icillin resistance in staphylococci,

vancomycin resistance in enterococci, and rifampin resistance in M. tuberculosis provides

examples of where molecular methods are used to supplement the growth-based methods

(165). However, it is difficult to imagine, given our current state of knowledge of the

molecular genetics of antimicrobial resistance and the technological limitations, that a

genotypic approach to routine antimicrobial susceptibility testing of bacteria could rival the

phenotypic methods in terms of information content and cost.

Molecular techniques are playing an increasing role in predicting and monitoring patient

response to antiviral therapy. The laboratory may have a role in predicting response to

therapy by detecting specific drug resistance mutations, determining viral load, and

genotyping. Both viral load and genotype are independent predictors of response to

combination therapy with pegylated interferon and ribavirin in chronic HCV infections,

although genotype is the main predictor of response (38, 50, 99, 187). Those patients with

high pretreatment viral loads (≥2 million copies/ml or 600,000 IU/ml) or genotype 1

infections have lower sustained response rates than do those with genotype 2 and 3

infections (38, 50, 99). Genotype is also used to determine the duration of therapy, with

genotype 1 infections requiring a longer course of therapy than genotype 2 or 3 infections

(28, 187). Recent studies have more closely defined duration of therapy based on the extent

of the viral response. Patients that do not reach a ≥2-log10 drop in viral load at 12 weeks

after initiating therapy are very unlikely to respond to pegylated interferon and ribavirin

(41). Moreover, patients with a rapid virologic response (HCV RNA level of <50 IU/ml 4

weeks after initiating therapy) may require a shorter duration of therapy, provided they have

a low baseline HCV RNA level (≤400,000 IU/ml) and minimal hepatic fibrosis (187).

Quantitative tests for HIV-1 RNA are the standard of practice for guiding clinicians in

initiating, monitoring, and changing antiretroviral therapy. Several commercially available

HIV-1 viral load assays have been FDA approved, and guidelines for their use in clinical

practice have been published (54; DHHS Panel on Antiretroviral

Guidelines, AIDSinfo.nih.gov). Viral load assays have also been used in monitoring response

to therapy in patients chronically infected with HBV (89) and in predicting the risk for

developing BKV-associated nephropathy in renal transplant recipients (60). In organ

transplant recipients, the persistence of CMV viral load after several weeks of antiviral

therapy is associated with the development of resistant virus (11).

LABORATORY PRACTICE

The unparalleled analytical sensitivity of nucleic acid amplification techniques coupled with

their susceptibility to cross contamination presents unique challenges to the routine

application of these techniques in the clinical laboratory. There are special concerns in the

areas of specimen processing, work flow, quality assurance, and interpretation of test

results. Additional information can be found in CLSI documents MM3-A2, Molecular

Diagnostic Methods for Infectious Diseases; Approved Guideline, 2nd ed. (22); MM6-

A, Quantitative Molecular Methods for Infectious Diseases; Approved Guideline (111); and

MM13-A, Collection, Transport, Preparation, and Storage of Specimens and Samples for

Molecular Methods; Approved Guideline (19).

Specimen Collection, Transport, and Processing

Proper collection, transport, and processing of clinical specimens are essential to ensure

reliable results from molecular assays. Nucleic acid integrity must be maintained throughout

these processes. Important issues to consider in specimen collection are the timing of

specimen collection in relationship to disease state and the proper specimen type. Other

factors that come into play include the use of the proper anticoagulant, transport and

storage temperatures, and time to processing of the specimen. HIV-1 viral load testing is an

example in which the proper conditions for specimen collection, transport, and processing

have been well described and has provided insight into the importance of these factors. For

HIV-1 viral load testing, the plasma needs to be separated from the cells within 6 h of

collection to minimize degradation of RNA. Once the plasma has been separated, it can be

stored at 4°C for several days, but −70°C is recommended for long-term storage (139).

Most types of specimens are best stored at −20 to −70°C prior to processing.

Molecular methods have several advantages over conventional culture with regard to

specimen collection. It may be easier to maintain the integrity of nucleic acid than the

viability of an organism. Molecular tests for the detection of C. trachomatis and N.

gonorrhoeae are an example in which DNA is stable on dry cervical swabs for a week at room

temperature or refrigeration temperatures, which is in stark contrast to the conditions

required to maintain organism viability for culture. Nucleic acid persists in specimens after

initiation of treatment (41, 83), thus allowing detection of a pathogen even though the

organism can no longer be cultured. Also, due to the increased sensitivity of molecular

assays, it may be possible to test a smaller volume of specimen or use a specimen that is

collected using a less invasive method.

The major goals of specimen processing are to release nucleic acid from the organism,

maintain the integrity of the nucleic acid, render the sample noninfectious, remove inhibiting

substances, and, in some instances, concentrate the specimen. These processes need to be

balanced with minimizing manipulation of the specimen. Complex specimen processing

methods are time-consuming and may lead to the loss of target nucleic acid or result in

contamination between specimens. Care must be taken to avoid carrying over inhibitory

substances, such as phenol or alcohol, from the nucleic acid isolation step to the

amplification reaction.

There are several general methods for nucleic acid extraction. Different methods may be

used depending on whether the desire is to purify RNA or DNA or both. Another factor to

consider when deciding on a nucleic acid extraction method is the type of pathogen sought.

Some pathogens, such as viruses, can be very easy to lyse, while mycobacteria,

staphylococci, and fungi can be very difficult to lyse. Enzyme digestion, harsh lysis

conditions, or mechanical disruption may be required to disrupt the cell walls of these

organisms.

DNA isolation methods often use detergents to solubilize the cell wall or membranes, a

proteolytic enzyme (such as proteinase K) to digest proteins, and EDTA to chelate divalent

cations needed for nuclease activity (6,47). The lysate can be used directly in amplification

assays, or additional steps may follow to purify the nucleic acid. These additional steps

remove proteins and traces of organic solvents and concentrate the specimen. In order to

successfully use a crude lysate, the target DNA must be present in a relatively high

concentration and there must be minimal inhibitors of amplification in the sample. If these

criteria are not met, additional purification steps should be used.

Another commonly used method of nucleic acid isolation involves disruption of cells or

organisms with the chaotropic agent guanidinium thiocyanate and a detergent (15). After a

short incubation, the nucleic acid can be precipitated with isopropanol. Guanidinium

thiocyanate denatures proteins and is also a strong inhibitor of ribonucleases, making it a

very useful tool for RNA isolation, although it is also used for purification of DNA. The Boom

extraction method is also based on the lysing and nuclease-inactivating properties of

guanidinium thiocyanate but utilizes the acid-binding properties of silica or glass particles to

purify nucleic acid (7). Over the past several years, various manufacturers have developed

commercially available reagents using one of these basic methods or a modification of these

methods. Many of these methods rely on the use of spin column technology, are easy to use,

and provide a rapid, reproducible method for purification of nucleic acid from a wide variety

of clinical specimens. In recent years, further advances have been made with the

introduction of magnetic silica particles which are coupled with instruments providing various

degrees of automation, thus further simplifying nucleic acid extraction and purification.

These reagents tend to be expensive, but the additional cost can be offset by labor savings.

Laboratories are increasingly using automated systems for nucleic acid extraction, as they

require less hands-on time, may reduce the risk of cross contamination between specimens,

and provide more consistent yields. There are now many automated systems available for

use in clinical laboratories; they should be thoroughly evaluated because not all isolate

nucleic acids with the same efficiency and purity. The quality of the nucleic acid can have a

significant impact on the performance of a molecular test.

Tissue samples need to be disrupted prior to the nucleic acid extraction process. This can be

accomplished by cutting the tissue into small pieces or mechanically homogenizing the tissue

prior to proceeding with one of the above-described extraction methods. Preserved tissue

specimens require removal of the paraffin with solvents and slicing into fine sections prior to

processing.

Removing inhibitors of amplification is a key function of the nucleic acid extraction process.

Simple methods of nucleic acid extraction that involve boiling of the specimen have been

used for relatively acellular specimens such as cerebrospinal fluid (CSF). Though the boiling

method is fast and easy, there are problems with inhibitors of amplification in CSF that are

not inactivated by boiling (104). The inhibition rate can be reduced to <1% by using a silicabased

extraction method. Similarly, crude lysates of urine and cervical swab specimens are

commonly used for the detection of C. trachomatis and N. gonorrhoeae. Specimens

containing amplification inhibitors have been reported to range from 1 to 5% for urine to as

much as 20% for cervical swabs (137). Common inhibitory substances include hemoglobin,

crystals, β-human chorionic gonadotropin, and nitrates. Blood samples are used commonly

for detection and/or quantification of a variety of viral pathogens, including HIV-1, HCV, and

CMV. HIV-1 viral load testing is an example in which the effects of different anticoagulants

have been well studied. HIV-1 viral RNA is most stable when collected in EDTA, and heparin

has been shown to be inhibitory to amplification and should be avoided (5, 67). In addition,

very small volumes of whole blood (1%) can be inhibitory to Taq DNA polymerase (58).

Other compounds such as acidic polysaccharides, which are components of glycoproteins

present in sputum and cervical specimens and bile salts found in stool, can also inhibit

polymerase (39). Human DNA, when present in the sample in high quantities, for example,

tissue or blood, may also interfere with the detection of a low concentration of pathogen

nucleic acid. With the recognition of such a wide array of inhibitors of amplification and the

availability of simple, reliable, semiautomated and automated nucleic acid extraction

methods, the use of crude lysates for testing becomes more difficult to justify. Regardless of

the nucleic acid extraction method employed, the laboratory should monitor inhibition rates

for different specimen types and nucleic acid extraction methods (see “Quality Control and

Assurance” below).

Contamination Control

Several types of contamination can occur with molecular testing: cross contamination of

specimens during the nucleic acid extraction step, contamination of specimens with positive

control material, and carryover contamination of amplified products. Contamination with

amplified products can occur with DNA or RNA target amplification and with probe

amplification methods. It does not occur with signal amplification assays, since nucleic acid

molecules are not synthesized with these methods. Cross contamination that occurs during

specimen processing or handling of positive control material can occur with all amplification

methods. The approach to the control of contamination due to amplified products has

changed dramatically with the widespread use of real-time amplification and detection

methods. Since the reaction tube is not opened after amplification, there is minimal risk of

contamination from the amplified product. Many laboratories using real-time methods

continue to use a variety of good laboratory practices to control for contamination, but the

focus is on minimizing cross contamination between specimens rather than contamination

from the amplified product. Refer to CLSI document MM3-A2, Molecular Diagnostic Methods

for Infectious Diseases; Approved Guideline, 2nd ed. (22), and Molecular Microbiology:

Diagnostic Principles and Practice, 2nd ed. (105), for detailed descriptions of good laboratory

practices to minimize contamination.

Clinical microbiologists have long been concerned about minimizing contamination between

samples with microorganisms during specimen processing. Molecular methods have raised

the level of concern considerably, and for good reason, as current methods can detect a few

molecules. The previously undetected low levels of contamination that occurred in processing

specimens for routine culture can lead to false-positive results in molecular assays.

Prevention of contamination due to target DNA or RNA is best done by careful handling of

specimens to avoid splashing, opening only one specimen tube at a time, pulse-spinning

tubes prior to opening, using screw-top tubes rather than snap-cap tubes to minimize

aerosolization, bleaching work surfaces, and using plugged pipette tips. Some of these

approaches can be difficult for high-volume laboratories, which is why automated extraction

systems can be very useful. Care must be taken with these systems to ensure that there is

no cross contamination during the automated process. This is often done by alternating

negative and high-titer specimens in a checkerboard arrangement and monitoring for

carryover of sample into the negative specimens. These experiments should be designed

with an understanding of the concentration of the organism in the clinical specimen. For

example, the concentration of HSV in CSF from patients with meningitis is quite low

compared to the concentration of BKV in the urine of a patient with nephropathy.

Preventing contamination of the laboratory with DNA from a clinical specimen or positive

control material is very important, because eliminating contamination with target DNA once

it occurs can be very difficult. This is why care should be taken to use a positive control at

the lowest concentration that consistently amplifies. The enzymatic and photochemical

inactivation methods used to control carryover contamination of amplified products are not

effective in preventing contamination with target DNA.

Enzymatic inactivation of amplified product can be accomplished with uracil-N-glycosylase

(UNG), a DNA repair enzyme found in a variety of bacterial species. During the PCR, dTTP is

replaced with dUTP so that dUTP is incorporated into the newly synthesized DNA products.

This allows for a distinction between starting template DNA and amplified products; only

newly synthesized PCR products will contain deoxyuracil. If UTP-containing amplification

products are present as contaminants, the addition of UNG to the reaction mixture will result

in the cleavage of deoxyuracil residues, thus destroying the contaminating DNA (95). The

use of UNG increases the amount of carryover DNA needed to contaminate the reaction

mixture by several orders of magnitude (124). When UNG is used, it is important to keep the

annealing temperature above 55°C so that the UNG remains inactive, thus avoiding

degradation of newly synthesized product. For the same reason, after completion of

amplification, the reaction mixture should be held at 72°C (168). UNG can be inactivated at

94°C, but prolonged inactivation at 94°C may also affect the activity of the polymerase

enzyme. UNG will not remove uracil from RNA molecules and is therefore ineffective in

controlling contamination in RNA amplification assays, such as TMA and NASBA.

When UTP and UNG are used, the PCR conditions should be reoptimized, as the magnesium

requirement may increase. The efficiency of amplification may be reduced when UTP is

substituted for TTP. This can be overcome by adding a mixture of dUTP and dTTP into the

master mix. The efficiency of inactivation using UNG depends on the size of the amplified

product and its G+C content. Inactivation may not be effective with amplified products of

fewer than 100 bp, as maximum UNG efficiency requires the DNA molecule to be 150 bp

(34).

Contamination of laboratory work surfaces, equipment, reagents, and clothing of laboratory

personnel with previously amplified nucleic acid products is of particular concern for clinical

laboratories, since these products can accumulate over time with routine testing and can be

inadvertently transferred to subsequent assay reactions, resulting in false-positive test

results. To minimize the potential for such amplicon contamination and false-positive results,

laboratories performing molecular tests with target amplification methods were designed

traditionally to have physical separation of preamplification (i.e., reagent preparation and

sample processing), amplification-detection, and postamplification (i.e., DNA sequencing)

areas with separate ventilation systems. In addition to the use of dedicated rooms, biological

cabinets, and dead-air boxes for various processes involved in specimen testing, laboratories

have also typically employed a unidirectional work flow for the movement of specimens,

supplies, and personnel from preamplification to postamplification areas through each phase

of testing. The physical separation of pre- and postamplification activities and a

unidirectional work flow are particularly important for those laboratories performing

postamplification analyses in which the reaction vessel is opened and the amplicon

transferred to another vessel or device (e.g., sequencing or liquid bead microarrays). The

strict separation of pre- and postamplification areas is less important for laboratories using

real-time amplification methods, particularly those using fully automated systems that

perform nucleic acid extraction, amplification, and detection.