Section 4 -

Molecular Diagnostics for Mycobacterial Infections Gary W. Procop

The diagnosis of tuberculosis has always been a challenge to the clinician and the laboratorian alike. The clinical diagnosis of tuberculosis may be foremost in the differential diagnosis when signs, symptoms, and exposure history are typical. However, atypical presentations have sundry manifestations wherein the possibility of tuberculosis may not be strongly suspected. For example, patients with remote histories of exposure and solitary skeletal lesions may not be initially suspected to have skeletal tuberculosis, until histopathology, often at the time of frozen section analysis, demonstrates necrotizing granulomas. [1] Similarly, tuberculous meningitis may not be suspected upon the initial presentation of the patient. [2] In addition, tuberculosis in the immunocompromised host poses a significant challenge. The isolation and characterization of the Koch's bacillus, demonstrated that culture-based methods could be used to demonstrate the presence of this microorganism in the respiratory, urine, or tissue specimens of infected patients. These methods have been refined over the many decades, and the phenotypic profiles of M. tuberculosis and the non-tuberculosis mycobacteria are well described. Although these methods have served the medical community for many years, and continue to do so, they are time-consuming – given the slow growth of mycobacteria. In addition, atypical reactions do occur and sometimes time-consuming tests need to be repeated which significantly delays the time to identification. The first major advance in the molecular identification of M. tuberculosis and other mycobacteria was the development and wide spread use of signal-generating hybridization probes that targeted species or complex-specific signature sequences in the ribosomal RNA of the mycobacterial cells. [3, 4] This approach proved feasible, since there are many mycobacteria present in a positive culture and many copies of rRNA in each individual mycobacterial cell (i.e., abundant target for probe hybridization). This approach became the standard for rapid mycobacterial identification and remains so in many institutions. Although useful, there are limitations with this approach. These assays have sensitivity limitations that preclude there use directly on clinical specimens. In addition, they only provide a single "yes/no" answer. This is excellent if the isolate is identified (i.e., a "yes" is obtained). Otherwise, the laboratorian only knows what it is not, and has to utilize either additional probes, which are expensive, or employ slower, traditional phenotypic methods. When Sanger or dideoxynucleotide sequencing by termination became more user friendly and cost-effective, then this technology began to be used for the identification of cultivated mycobacteria. Hundreds of nucleotides of sequence may be generated by this method, following a "broad-range" PCR that amplifies the DNA segment of interest from the mycobacterium to be identified. [5, 6] The requirements for such a PCR assay is that the primers used should amplify all clinically important mycobacteria, whereas the sequences that are between the primers afford the differentiation of species or clinically-important complexes. Not surprisingly, the 16S rDNA gene is commonly used for this purpose, but other genetic targets, such as the rRNA polymerase gene (rpoB) have also been demonstrated to be useful for this purpose. More recently, a newer sequencing chemistry – sequencing by synthesis – or pyrosequencing – has been demonstrate to achieve similar differentiation, although only a limited (30-50) nucleotide sequences are typically generated. [7] DNA microarrays may also be used following a broad-range PCR. [8, 9] These technologies effectively achieve the simultaneous hybridization of multiple probes (even hundreds) in a single reaction. The signals may generate a fluorescent or, more recently, an electrical signal. The traditional limitations of these technologies have been the cost and the amount of data generated. These have been demonstrated to be able to identify the common, clinically-relevant mycobacteria, as well as the genetic determinants of resistance to rifampin and kanamycin. More recently, limited microarray have been designed, which use a smaller number of probes and therefore generate less data. These limitations in the size of the microarray will also have an influence on cost. The approaches heretofore described have had a marked influence on the time to identification once a mycobacterium species is recognized in culture, but the another approach has been to directly characterize clinical specimens for the presence or the absence of M. tuberculosis complex. [10, 11] These approaches, when used in conjunction with clinical skills and judgment, may optimize the use of respiratory isolation and direct therapy. Many M. tuberculosis complex-specific nucleic acid amplification assays have been developed, two of which have FDA approval. Additionally, assays have been developed that recognize M. tuberculosis complex, as well as some of the more commonly encountered non-tuberculosis mycobacteria. [12, 13] Molecular technologies and techniques have evolved significantly since the time when the first hybridization probes were used for the identification of mycobacteria. These methods, although often expensive, afford the possibility to rapidly identify cultivated mycobacteria, and to characterize clinical specimens with a reasonable degree of certainty for the presence or absence of M. tuberculosis. Regardless of the power of these techniques, they remain adjunct tools and are not (yet) replacements for the acid fast stain and culture. References Kobayashi, N., et al., The use of real-time polymerase chain reaction for rapid diagnosis of skeletal tuberculosis. Arch Pathol Lab Med, 2006. 130(7): p. 1053-6. Carrol, E.D., J.E. Clark, and A.J. Cant, Non-pulmonary tuberculosis. Paediatr Respir Rev, 2001. 2(2): p. 113-9. Ephraim, D.A. and E.D. Spitzer, Use of acridinium-ester-labeled DNA probes for identification of mycobacteria in Bactec 13A blood cultures. Diagn Microbiol Infect Dis, 1994. 18(3): p. 137-9. Middleton, A.M., M.V. Chadwick, and H. Gaya, Detection of Mycobacterium tuberculosis in mixed broth cultures using DNA probes. Clin Microbiol Infect, 1997. 3(6): p. 668-671. Roberts, G.D., E.C. Bottger, and L. Stockman, Methods for the rapid identification of mycobacterial species. Clin Lab Med, 1996. 16(3): p. 603-15. Han, X.Y., et al., Rapid and accurate identification of mycobacteria by sequencing hypervariable regions of the 16S ribosomal RNA gene. Am J Clin Pathol, 2002. 118(5): p. 796-801. Tuohy, M.J., et al., Pyrosequencing as a tool for the identification of common isolates of Mycobacterium sp. Diagn Microbiol Infect Dis, 2005. 51(4): p. 245-50. Troesch, A., et al., Mycobacterium species identification and rifampin resistance testing with high-density DNA probe arrays. J Clin Microbiol, 1999. 37(1): p. 49-55. Caoili, J.C., et al., Evaluation of the TB-Biochip Oligonucleotide Microarray System for Rapid Detection of Rifampin Resistance in Mycobacterium tuberculosis. J Clin Microbiol, 2006. 44(7): p. 2378-81. Abe, C., et al., Detection of Mycobacterium tuberculosis in clinical specimens by polymerase chain reaction and Gen-Probe Amplified Mycobacterium Tuberculosis Direct Test. J Clin Microbiol, 1993. 31(12): p. 3270-4. Nolte, F.S., et al., Direct detection of Mycobacterium tuberculosis in sputum by polymerase chain reaction and DNA hybridization. J Clin Microbiol, 1993. 31(7): p. 1777-82. Shrestha, N.K., et al., Detection and differentiation of Mycobacterium tuberculosis and nontuberculous mycobacterial isolates by real-time PCR. J Clin Microbiol, 2003. 41(11): p. 5121-6. Cormican, M., et al., Multiplex PCR for identifying mycobacterial isolates. J Clin Pathol, 1995. 48(3): p. 203-5.