| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
Institute of Medical Microbiology, and Hygiene, University Hospital of Regensburg, Franz-Josef-Strauss-Allee 11, D-93053 Regensburg, Germany, E-mail udo.reischl{at}klinik.uni-regensburg.de, Fax 49-941-944-6402
Effective treatment of bacterial infections ideally requires the rapid and accurate detection and identification of the causative organism. Among the procedures currently used in clinical microbiology laboratories, the most precise method of detecting bacteria is growth in a culture. Although at least 8 h of incubation are required, diagnostic culture indicates the presence of living and therefore potentially infectious organisms and provides the material for subsequent antibiotic susceptibility testing.
Starting with a pure culture or single colonies of the grown organisms, species identification is usually accomplished by a comprehensive determination of phenotypic profiles including Gram stain results, morphology, growth requirements, biochemical or biophysical properties, and when available, specific antigen or agglutination tests. This method has been automated over the past few years and has become a cost-effective way to identify the majority of bacterial pathogens. Because most of the phenotypic and biochemical properties are determined by the metabolism of actively growing cells, analysis requires a considerable number of pure organisms, and differentiation down to species level may require hours or days of subculture in the presence of various substrates and/or selective media. For slow-growing or nonreactive bacterial species, the results are often delayed. Moreover, accurate identification is compromised when common bacterial species present with uncommon phenotypes, or when unusual species are encountered whose phenotypic profiles are not yet covered by the database.
The advent of in vitro nucleic acid amplification and, more recently, the introduction of rapid-cycle real-time PCR technology has led to substantial improvements in the diagnosis of infectious disease (1)(2). By eliminating the need for subculturing the target organisms before their definitive identification, nucleic acid amplification– based procedures can provide highly desired same-day results, and genotypic methods have evolved as objective, rapid, and accurate ways for bacterial species identification in modern clinical microbiology. Driven by a continually increasing demand for rapid test results that impact clinical outcome, and facilitated by the steadily rising performance of technology, the number of applications for detecting bacterial, viral, or parasitic pathogens, resistance genes, or genetically encoded pathogenicity factors is growing exponentially (1).
The clinical presentations of most infectious diseases are associated with a list of pathogens under suspicion that can be tested for in a selective or specific manner. Alternatively, a so-called eubacterial, broad-range, or panbacterial analysis can be performed. In assays for species or pathogenicity factor identification, characteristic marker genes or chromosomal segments of cryptic function can be used as unique target sequences. In contrast, panbacterial approaches are derived from certain genes or regions of the bacterial chromosome that consist of species-specific segments flanked by widely conserved sequences, allowing for efficient annealing of panbacterial primers (3). Among bacterial genes, the gene encoding 16S rRNA (16S rDNA) is one of the most favorable candidates.
The primary, secondary, and tertiary structures of the rRNA molecules are vital for cellular growth, function, and survival. Consequently, the corresponding genes are among the most conserved molecules in bacteria and other organisms. Conservation within the 16S rDNA is such that some sequences are shared by all bacteria, some by species of the same genus, and some regions are species specific. Because 16S rDNA has played a seminal role in evolutionary and phylogenetic studies over the last 20 years, it has become a widely accepted species marker in diagnostic as well as systematic bacteriology. Reference databases of such sequences have been growing rapidly (4)(5)(6)(7), and data quality has been improved with submission of validated complete 16S rDNA sequences (8), thus promoting this approach as a new gold standard for broad-range microbial identification. Several alternative, broadly conserved, and phylogenetically useful gene sequences have also been explored as bacterial species markers, including 23S rDNA, internal transcribed spacer regions, hsp65, rpoB, recA, sodA, gyrB, IF2, and EF1A. Once the sequence region of interest is amplified from crude or purified lysates of the cultured bacterial organisms, the sequence of the amplicons contains the information necessary to identify the underlying species.
Since the late 1980s, when we enthusiastically but arduously read the base ladders of 32P-labeled fragments on x-ray films, DNA sequencing technology has made tremendous progress. Recent advances include capillary and microelectrophoretic separations, sequencing by hybridization, and cyclic array sequencing (9)(10). However, it is still beyond the means of many clinical microbiology laboratories to sequence a large number of microorganisms on a routine basis. Alternative strategies to characterize amplicons at the nucleotide sequence level include solid-phase hybridization on high-density arrays (11) and electrospray ionization mass spectroscopy (12). Although recent advances in these technologies have decreased the size and increased both throughput and user-friendliness, the devices are still several steps away from easy implementation in the routine clinical microbiology laboratory.
The report of Cheng et al. (13) in this issue of Clinical Chemistry makes complex solutions for amplicon characterization appear to be "technical overkill" with respect to rapid bacterial species identification. Starting with a pure culture of bacterial organisms, this novel approach relies on panbacterial real-time PCR followed by high-resolution melting analysis (14), thus offering a simple workflow and overall result turnaround times of 
1.5 h without the need for laborious and contamination-prone postPCR manipulations or expensive DNA sequencing equipment.
High-resolution melting analysis relies on a central principle of molecular diagnostics: the Watson-Crick base pairing of double-stranded DNA, for which the melting temperature (Tm) depends on the number, type, and succession of the component base pairs. At the simplest level, the 3 hydrogen bonds and stacking energy of a G-C base pair contribute more to stability than an A-T base pair. To a first approximation, Tm can be estimated by length and GC content. Better estimates of Tm are obtained by considering the nearest neighbors of each base pair (15). Even when the length and GC content are identical, slight differences in the nucleotide sequence often give rise to different Tms of the respective double-stranded DNA molecules. Of course, Tm represents only 1 point on a DNA melting curve, and when the entire curve is considered, even more sequence discrimination is possible. Thus, a precisely determined melting curve is very characteristic of a given double-stranded DNA sequence.
A 200 nucleotide DNA strand has 4200 different possible sequences, but the Tm varies over only a 40 °C range. Because of the relatively high proportion of conserved segments in ribosomal genes of bacterial organisms, the maximum impact of genus- and species-specific signature sequences on the Tm of a corresponding 200 bp amplicon is typically within a +/– 3 °C range. With the availability of high-resolution melting instruments, melting curves can be measured with a precision of a fraction of a °C. Once species-specific "melting curve signatures" of PCR amplicons have been determined and deposited in a high-resolution melting database, the melting curves of amplicons generated from bacterial organisms of unknown identity can be assigned. And, having the complex eubacterial phylogenetic tree in mind, a precisely determined melting curve should allow at least for a reliable differentiation between members of the major branches. According to the data of Chen et al. (13), Tm-based differentiation is possible down to genus or species level. Within a group of 25 bacterial species frequently encountered in positive blood cultures, 9 species presented with unique Tms, and the remaining 16 species could be classified into 4 distinct "melting groups".
As with every forward-looking diagnostic approach, there are limitations as well as advantages. Whereas 16S rDNA sequences are highly conserved, the genome of some bacterial species harbors multiple copies of the ribosomal operon. Any nucleotide polymorphisms between the intragenomic 16S rDNA copies may cause difficulties in obtaining interpretable melting curves (16)(17). Moreover, it is well known that certain bacterial species share almost identical 16S rDNA sequences (e.g., E. coli, Salmonella and Shigella). In such cases, more discriminatory regions are necessary (18), or a battery of species-specific assays can be performed.
Another inherent dilemma of this biophysical approach is that some bacterial species with amplicons of different sequence have almost identical Tms. To resolve such "melting groups", Cheng et al. (13) performed several subsequent heteroduplex-melting analyses with amplicons of the unassigned bacteria and amplicons derived from a reference strain. When amplicons of a given species are cohybridized with amplicons of a defined type strain, any sequence variation between them is easy to observe because some mismatched hybrids (heteroduplexes) are formed that alter the shape of the melting transition (19). This "heteroduplex effect" has a much greater influence on the shape of the melting curve than on its absolute Tm. The ability to observe the entire melting curve adds another range of analytical possibilities to high-resolution melting.
DNA sequencing will keep its central role for detailed characterization of bacterial marker genes in systematic microbiology. However, Cheng et al. (13) have demonstrated the feasibility of using high-precision melting analysis for bacterial speciation. This will be of considerable interest to clinical microbiologists seeking a rapid molecular method for species identification from positive blood cultures without sophisticated sequencing equipment. Because an early clue about the underlying genus or species may have a major impact on patient management and antimicrobial therapy, the eagerly awaited results of microscopy and Gram staining might be complemented by the nearly isochronous but more discriminative results of eubacterial PCR and high-resolution melting. Although it is not a turnkey solution to complete bacterial species identification, the study of Cheng et al. (13) will inspire continued progress toward simple solutions for amplicon characterization and differentiation for species identification in the microbiology laboratory.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J.-H. Lin, C.-P. Tseng, Y.-J. Chen, C.-Y. Lin, S.-S. Chang, H.-S. Wu, and J.-C. Cheng Rapid Differentiation of Influenza A Virus Subtypes and Genetic Screening for Virus Variants by High-Resolution Melting Analysis J. Clin. Microbiol., March 1, 2008; 46(3): 1090 - 1097. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
