Technology Comparison

Current molecular typing methodologies, including manual rep-PCR, have limitations (Versalovic and Lupski, 1996; van Belkum et al., 2001; van Belkum, 2003; Zaidi et al., 2003). According to a review by Soll published in 2000 (Soll, 2000), at that time a comprehensive strain-typing method was not available because no typing method was practical for assessing contamination in real-time (Woods et al., 1996; Shopsin and Kreiswirth, 2001), for providing complete tracking data (Dahl et al., 1999; Suppola et al., 1999), or had data archiving capability, all of which were required to build libraries and share data among laboratories. The DiversiLab System was specifically designed to integrate these components into a commercially available package (Healy et al. 2005). Table 26.1 provides a comprehensive comparison of many commonly used strain-typing methods.

Discrimination to the subspecies or strain level is essential to a clinical strain-typing system because many organisms, such as methicillin-resistant Staphylo-coccus aureus (MRSA), are extremely heterogeneous. Although some typing systems may be valid for more diverse organisms, many fail to discriminate at the strain level for these clinically important organisms. For example, ribotyping often has difficulty distinguishing among different subtypes (Kostman et al., 1995; Lobato et al., 1998), and 16S rRNA sequencing often shows low levels of subspecies and strain discrimination (Sander et al., 1998; Barney et al., 2001; Callon et al., 2004). Additionally, plasmid typing and MLEE provide only an estimate of genetic relatedness (Mayer, 1988; Seifert et al., 1994a, 1994b; Trindade et al., 2003). The DiversiLab System provides genus-specific kits that have been developed for a high level of strain discrimination. However, universal kits, such as the DiversiLab Gram-negative kit, can be used when speed is more important than discrimination. Even when the genus or species identification is inconclusive, the DiversiLab System can reveal relationships among samples and thus help identify the contaminant source. PFGE is considered the gold standard because of its high discriminatory power (Tenover et al., 1995), and RAPD is also highly discriminating; however, both methods show poor interlaboratory reproducibility (Davis et al., 2003). Table 26.2 is a summary of the organism-specific applications of each technology.

Table 26.1. Comparison of strain-typing technologies.

Ease of Ease of

Table 26.1. Comparison of strain-typing technologies.

Ease of Ease of

Technique

performance0

interpretation0

Discrimination0

Time0

Reproducibility0

Cost0

Comercial kit

PFGE

Moderate

Moderate-High

High

High

High

Moderate-High

No

Sequencing

Low

Moderate-High

High

Moderate

High

High

No

RAPD

High

High

High

Low

Moderate

Low

No

RFLP

Moderate

High

Moderate

Moderate

High

Low

No

AFLP

Moderate

Moderate

High

Moderate

Moderate

Moderate-High

Available'

Ribotyping

Moderate

Moderate

Moderate

Moderate

High

High

Available'

MLST

Low

Moderate-High

High

Moderate

Highe

Highe

No

Rep-PCR

High

High

High

Low

High

Low

Available'

PFGE, pulsed-field gel electrophoresis; RAPD, randomly amplified polymorphic DNA; RFLP, restriction fragment length polymorphism; AFLP, amplified fragment length polymorphism; MLST, multilocus sequence typing.

0 Table 1 (Olive and Bean, 1999 J Clin Micribiol, 37(6):1661-1669 and/or Table 2 (Van Belkum, et al., 2001 Clin Microbilol Rev, 14(3):547-560). b Applied Biosystems Microbial AFLP Kits. c DuPont RiboPrinter. d Bacterial Barcodes Diversilab System. e Trinidade, et al. 2003 Braz J Infect Dis, 7(l):32-43.

Table 26.2. Organism-specific applications of each technology.

Filamentous

Protocol consistency

Technique

Bacteriaa

Mycobacterium1

fungi

Yeasta

between organisms

PFGE

Yes

Limited*

No

Yes

Low, RE and electrophoresis parameters change with each organism*

Sequencing

Yes

Yes

Yes

Yes

Moderate, species-specific gene-related assays required for strain-typingc

RAPD

Yes

Yes

Yes

Yes

Low, very sensitive to primer and annealing temperature changesc

RFLP

Yes

Yes

Yes

Yes

Moderate, gene-specific primers change with each organismc

AFLP

Yes

Yes

Yes

Yes

Moderate, nonspecific primer kits availabled

Ribotyping

Yes

Yes

No

No

Moderate, nonspecific restriction enzyme kits availablee

MLST

Yes

No

Limited'

Yes

Low, multiple primers change with each organism f

Rep-PCR

Yes

Yes

Yes

Yes

High, quality-controlled genus-specific kits available®

PFGE, pulsed-field gel electrophoresis; RAPD, randomly amplified polymorphic DNA; RFLP, restriction fragment length polymorphism; AFLP, amplified fragment length polymorphism; MLST, multilocus sequence typing.

a Data taken from Table 4, Pfaller (2001) Emerg Infect Dis, 7(2):312-318. b http://www.cdc.gov/pulsenet/protocols.htm. c Olive and Bean (1999) J Clin Micribiol, 37(6):1661-1669. d http://www.appliedbiosystems.com. e http://www.qualicon.com/riboprinter.html. f Trinidade, et al. (2003) Braz J Infect Dis, 7(1):32-43. g http://www.bacterialbarcodes.com. h Zhang, J Clin Microbial, 2004; 42(12):5582-5587. ' Taylor, CurrOpin Microbiol, 2003; 6:351-356.

Reproducibility of a typing method is critical for longitudinal studies, including tracking and trending, for comparing archived fingerprint patterns, and for validation of the system. The data presented in several studies (Versalovic et al., 1992; Kang and Dunne, 2003) indicate that rep-PCR fingerprints are stable over multiple generations of growth, reproducible within a plate of isolated colonies within a strain, and distinct between strains. Additionally, automated rep-PCR has shown high interlaboratory reproducibility (Healy et al., 2005; Shutt et al., 2005).

Some sequencing techniques, including MLST, also show high reproducibility (Storms et al., 2002; Tavanti et al., 2003). Conversely, both automated ribotyping and AFLP have been shown to be reproducible only after manipulation of the fingerprint patterns to achieve high reproducibility (Bagley et al., 2001; Brisse et al., 2002).

Commercialization of a typing system can increase the reproducibility due to standardization of reagents and data collection. The DiversiLab System has been commercialized with the appropriate controls to facilitate quality-control efforts. Variation in the template DNA concentration, instruments, laboratory facilities, or operator do not affect the reproducibility of the assay, verifying the ease of use, portability of the data, and robustness of automated rep-PCR (Cangelosi et al., 2004; Healy et al., 2004; Healy et al., 2005; Shutt et al., 2005). All manufactured kits are quality controlled and include positive and negative controls. The DiversiLab System instrumentation comes with documentation of design qualification, including a Declaration of Conformity for manufacturing specifications and a CE mark. Additionally, installation of the system includes on-site training, a certification panel for technical performance, and qualification for laboratory thermal cyclers. Most typing methods must still be performed with "homebrew" reagents and nonstandardized protocols, although ribotyping has been commercialized as the DuPont RiboPrinter and AFLP has been commercialized in ready-made primer kits from Applied Biosystems to be used with their Genetic Analyzer and GeneMapper software.

Automation of a methodology often allows the technology to be more rapid, which is important for tracking infections in real-time, and easier, which is required for routine clinical labs. PFGE, RAPD, and AFLP have extensive technical hands-on time or require highly skilled technicians (Kostman et al., 1995; Tenover et al., 1995; Lobato et al., 1998; Olive and Bean, 1999; van Belkum et al., 2001; van Belkum, 2003; Zaidi et al., 2003). Although, MLST can be used as a non-culture-based typing method, it can be labor intensive and costly (Shopsin and Kreiswirth, 2001; Diep et al., 2003; van Belkum, 2003; Zaidi et al., 2003). Because much of the workflow is automated, the DiversiLab System overcomes these issues (Pounder et al., 2005; Shutt et al., 2005). However, the process still requires some manual manipulation of the samples. The current recommended DNA extraction method is the MoBio UltraClean Microbial Isolation Kit due to its high, consistent yield from Gram-positive, Gram-negative, and fungal isolates. Twenty-four samples can be processed in an hour; this method is simple, yet laborious (Shutt et al., 2005). To overcome the hands-on portion of this procedure, automated extraction methods can be coupled with the DiversiLab System; however, validation of these methods may be required.

Often, the downside of technology commercialization is the associated cost. Automated ribotyping is generally much more expensive than other typing techniques, including rep-PCR (Inglis, 2002; Silbert, 2004). Both noncommercial and commercial AFLP are expensive due to the requirement of a sequencer (Olive and Bean, 1999), which is also required for MLST and other sequencing protocols.

Noncommercial typing systems may also be expensive if they require specialized equipment, such as the specialized gel-electrophoresis unit required for PFGE. The DiversiLab System has a lower capital equipment cost and competitive cost per test.

One issue common to all fingerprinting methods, including the DiversiLab System, is the difficulty of the data interpretation, mainly because the results are qualitative and complex. For example, chromosomal RFLP and AFLP yields complex DNA profiles that can be challenging to interpret (Vos et al., 1995; Olive and Bean, 1999). Although standardized guidelines have been applied to PFGE (Tenover et al., 1995), this method is often time-consuming and difficult because it depends on visual analysis of the fingerprints. With the DiversiLab System software, interpretation continues to be subjective. However, an interpretation guide is provided and the software provides a variety of analysis tools to assist the user. To aid the infection control team in producing specialized data reports, such as a search on a particular isolate over a specified period, the DiversiLab software offers 10 fields to incorporate sample demographics, which are linked to the isolate and can be displayed on the report (Fig. 26.2). These demographics may include time, date, location, or other particulars about the sample. By using the DiversiLab software package to assist in monitoring over time and providing useful trend reports, proactive measures can be taken to prevent contamination.

Was this article helpful?

0 0

Post a comment