European Review for Medical and Pharmacological Sciences 2012; 1(Suppl 2): 73-78 Microbiological and immunological diagnosis of tuberculous G. DELOGU, A. ZUMBO, G. FADDA

Institute of Microbiology, School of Medicine, Catholic University of the Sacred Heart, Rome (Italy)

Abstract. – Background: Tuberculous one of the deadliest infectious disease agents, spondylodiscitis is one the many manifestations causing 1,7 million deaths and with 8,4 million of active (TB) and can result follow- new cases annually1. Most of the TB cases occur ing primary infection or, more frequently, from reactivation of active TB in subjects with latent in developing countries where TB poses a major TB. Definitive diagnosis of tuberculous spondy- health problem, with relevant social and econom- lodiscitis requires the identification of Mycobac- ic implications. terium tuberculosis in the biological sample fol- (M.) tuberculosis is a human lowing microbiological analysis. pathogen that belongs to the species Mycobacteri- Aim: To summarize the recent advancement um tuberculosis (MTB) complex, which compris- in the diagnosis of TB, focusing on classical and molecular microbiological procedures, providing es M. africanum, which causes TB in humans but an overview on the recent advancements in the is restricted to Africa, M. bovis, which is the etio- understanding of TB pathogenesis and their im- logic agent of TB in animals and M. microti plications for the immunological diagnosis which infects voles and does not infect humans2. Materials and Methods: Isolation in culture The complete sequencing of the M. tuberculosis of the bacilli and detection using molecular genome has provided a new understanding on the tools are the gold standards, though sensitivity biology of the tubercle bacillus, leading to a bet- of these assays is significantly lower compared to what observed for pulmonary TB, making di- ter understanding of the mechanism of pathogen- agnosis of spinal TB challenging. esis and the definition of the genetic determinants Results: The use of the interferon-γ release of drug resistance, opening new avenues to com- assays (IGRAs) for the immunological diagnosis bat the disease through more effective prophylac- of TB infection could be of help and shall pre- tic and therapeutic measures3-5. The 80 years old cede the invasive techniques, such as or live and attenuated vaccine strain Bacillus of Cal- , required to obtain the biological sample. IGRAs measure the presence of effector T cells mette and Guerin (BCG), while ineffective in pre- in the blood that can readily respond to an anti- venting TB in adults, remains the only tool avail- genic stimuli by secreting cytokines, and that are able to reduce the risk of fatal and systemic TB an indication of the presence of the bacilli in disease in children. There is an urgent need for a vivo. IGRAs are more sensitive and specific than new and improved vaccine against TB and many the intradermic reaction of Mantoux, though both efforts are underway using traditional and highly these immunological tests cannot distinguish be- innovative strategies6,7. The same is true for the tween latent TB infection and active TB. Conclusions: A modern diagnosis of TB therapeutic regimens that still rely on more than spondylodiscitis should rely on the use of mi- forty years old drugs such as , rifampin, crobiological and immunological assays and the , , and few latter could potentially be of great help in moni- other second line drugs8. toring therapy effectiveness. The new interest in TB research, that coincid- ed with the spread of the HIV pandemic and the Key words: re-emergence of TB in western countries, had an impact primarily on the diagnosis of TB, with the Tuberculosis, Spinal TB, Pott’s Disease. introduction of more sensitive, specific and rapid methods that have contributed significantly to the control of TB in the last ten years. Introduction Microbiological diagnosis of TB is always a challenging task, particularly when non-pul- TB is one of the ancient human scourge that monary TB is suspected, as in the case of has afflicted mankind since millennia and is still spondylodiscitis9,10, since the biological samples

Corresponding Author: Giovanni Delogu, PhD; email [email protected] 73 G. Delogu, A. Zumbo, G. Fadda available to detect M. tuberculosis are difficult to specific cell mediated immune response specific obtain. At the same time, several causative agents for M. tuberculosis, classically highlighted by the of infectious spondylodiscitis have been identi- Mantoux test16. Latent TB can last for decades fied, which require different antimicrobial thera- and people with latent TB have a ≅ 5% chance of pies. For these reasons, it is recommended that developing active TB during lifetime, indicating the start of the antimicrobial therapy in case of that the host immune response cannot completely infectious spondylodiscitis should be delayed un- eradicate the bacteria in vivo. These observations til a microbial diagnosis is established10. have puzzled scientists since the early 20th centu- In this review, we will summarize the recent ry and only recently we are gaining a new under- advancement in the diagnosis of TB, focusing on standing of the immunological processes associat- classical and molecular microbiological proce- ed with latent TB and on the metabolic status of dures, providing an overview on the recent ad- the bacteria during this stage17,18. The old view vancements in the understanding of TB patho- that latent M. tuberculosis exists in classic TB le- genesis and their implications for the immuno- sions such as the primary complex has been con- logical diagnosis and usefulness that these new tradicted by many findings, including the demon- tests may have in guiding the diagnostic query. stration that bacilli can be found in non-profes- sional phagocytic cells such as endothelial cells, pneumocytes and fibroblast that are scattered in the lung tissue19. M. tuberculosis also persists in Pathogenesis of Tuberculosis adipose tissue where it can remain in a non-repli- cating state avoiding killing by antimicrobials and TB is an airborne transmitted disease, with the recognition by the effector immune cells20.The bacilli released by a patient with active pul- status of the bacilli during latent TB may vary monary TB inhaled by a healthy person11.The greatly from a persistence non-replicating state, to bacilli once in the alveoli are ingested by dormancy, to active replication, and M. tuberculo- macrophages which can kill the bacteria, though sis can either use aerobic and anaerobic metabo- in 20-50% of the exposed subjects, M. tuberculo- lism16-18. What is remarkable is that M. tuberculo- sis resists to the innate immune response and ac- sis can resist the attack from many immune cells, tively multiply within the macrophages. Nearby including CD4 and CD8 T cells, which provide cells are infected by the replicating bacilli, which an harsh and oxidizing environment for the bacte- cause a local inflammation that activates the sec- ria21. In line with these findings, Fortune et al22 ondary host immune response12,13. It is at these have recently demonstrated that the mutation rate early times of the infectious process, when the of M. tuberculosis during latent TB is similar to host cannot yet contain bacterial multiplication, that observed during active TB, as a results of the that M. tuberculosis spreads through the lym- active bacilli replication and of the oxidative phatics, but most importantly can translocate to damage to the M. tuberculosis DNA exerted by the blood stream and spread haematogenously to the antimicrobials produced by activated potentially reach any organ14,15. The mounting ac- macrophages. quired cell mediated immune response then be- As a result of these findings, the traditional di- gins to contain the infection, starting at the site of chotomy of TB status (latent TB and active TB) primary infection when the cellular infiltrate or- is being revisited, with the introduction of the ganize in the typical granuloma. Only in ≅ 5% of concept of TB spectrum, which identify a dy- cases, the host immune response fails to control namic equilibrium between the host immune re- primary infection which then progresses to overt sponse and the bacilli16,18. The TB spectrum is pulmonary TB and it is the immune response it- seen as a gradient that may potentially vary from self that is responsible for the extensive tissue latent TB in an almost dormant state to overt ac- damage and necrosis which is the hallmark of ac- tive TB, and any event that can affect this equi- tive TB in immune-competent patients11. librium may lead to the progression of latent TB Conversely, it is estimated that in 90-95% of to overt disease. infected subjects, the host immune response can Hence, tuberculous spondylodiscitis, also contain bacterial multiplication, leading to latent called Pott’s disease, occurs when the bacteria infection (latent TB). Latent TB is clinically that has reached the spine, starts replicating in the silent, with no outward signs or symptoms of dis- tissue forming a secondary lesion that progresses ease, and is characterized by the presence of a to cause the typical immunopathogenesis and tis-

74 Microbiological and immunological diagnosis of tuberculous spondylodiscitis sue damage that may evolve in complications identification of the epidemiological features of such as vertebral collapse, and the M. tuberculosis strain. Unfortunately, M. tu- paraspinal . Since haematogenous berculosis is a slow growing mycobacteria and spread of the bacilli occurs readily following pri- isolation in culture may require up to 45 days, re- mary infection in the lung tissue, spinal TB can sulting in a delay of diagnosis and in the initia- develop either soon after primary infections or tion of appropriate therapy. many years or decades later following reactiva- A recent report by Kumar et al24, where 51 pa- tion. Moreover, since during latent TB a dynamic tients with Pott’s disease were evaluated, indicat- equilibrium between the host immune response ed that the sensitivity of the microscopic analysis and actively replicating bacilli is maintained, and was only of 33%, that of culture of 43% and the bacteria can be found in many tissues and cells, two combined of 59%. These results underscore any event that can suppress the immune response the problems associated with the microbiological can results in disease reactivation. For these rea- diagnosis of TB . sons, tuberculous spondylodiscitis can occur in The use of molecular diagnosis to detect and patients with concurrent infections such as HIV identify M. tuberculosis directly in the specimen or undergoing biological therapy with anti-TNF is of great usefulness, given its high sensitivity drugs. Osteoarticular TB can also be associated and the possibility to obtain the results in one day. with TB disease elsewhere in the body including There are several commercially available assays pulmonary TB. In any case, proper diagnosis of to detect M. tuberculosis in biological specimens spinal TB is a challenging task for the clinician and though these tests have not yet been licensed that must identify the proper therapeutic regimen. for use in non-respiratory specimens, several re- ports clearly indicate that they should be consid- ered highly reliable25,26. The specimens sent to the laboratory are directly subjected to amplification Direct diagnosis of specific M. tuberculosis DNA sequences using one of the many available techniques such as Definitive diagnosis of TB requires the detec- polymerase chain reaction (PCR), real-time PCR tion of M. tuberculosis from the biological sam- or strand displacement amplification. To broaden ple by at least one of current microbiological the search for the detection of other possible bac- techniques: microscopical analysis, isolation in terial infectious agents, it is possible to amplify culture or molecular methods. There are several by PCR the DNA genomic fragment encoding the major problems with the direct diagnosis of TB 16S RNA fragment, using universal primers that spondylodiscitis. First, the biological sample can can amplify DNA from any bacteria. If an ampli- be obtained only using invasive techniques, such fication is achieved, the DNA fragment is subject- as biopsy or tissue sampling during surgery. Sec- ed to sequencing to identify the bacterial species ond, TB spondylodiscitis is usually paucibacil- detected in the specimen27. lary, which makes microscopical analysis follow- While it is widely accepted that molecular ing acid-fast staining not very reliable do to its tools can greatly improve the diagnosis of tuber- low sensitivity. Moreover, microscopic analysis culous spondylodiscitis, more studies are re- does not allows the proper identification of the quired to carefully determine the sensitivity and mycobacterial species. Isolation in culture of M. specificity of these assays also in comparison to tuberculosis from the specimen is considered the traditional microscopic and culture techniques. gold standard, since it demonstrate the presence of the etiologic agent and allows for the determi- nation of the susceptibility testing to first and second line drugs. This is becoming of para- Immunological diagnosis mount importance at a time when M. tuberculo- sis strains resistant to two more drugs currently The immunological diagnosis of TB has been used in TB therapy are found responsible for TB historically performed by the or tu- disease, including spondylodiscitis23. Moreover, berculin skin test (TST), which consists in the in- the isolation in culture is required to extract ge- tradermic inoculation of the purified protein de- nomic DNA that could be used for the identifica- rivative (PPD) and in measuring 48-72 hours lat- tion of the genetic determinants of drug resis- er a delayed-type hypersensitivity which measure tance and the genotypic characterization and the cell mediated immune response against the

75 G. Delogu, A. Zumbo, G. Fadda tubercle bacilli. In the last few years, more sensi- from blood drawn from subjects with latent TB tive and specific immunological assays designed respond to the HBHA stimuli by secreting inter- for the immunological diagnosis of M. tuberculo- feron-γ while patients with active TB are unable sis have been developed and are now widely used to deploy a similar response31,32. Interestingly, pa- as surrogate of TST28. These are named interfer- tients with active TB develop antibodies against on-γ release assays (IGRAs) and measure the in- HBHA while subjects with latent TB do not31-33. terferon-γ cytokine released by T cells obtained Recent advances in the purification protocol of from a blood sample following restimulation HBHA34 have made possible to use this protein with M. tuberculosis antigens. These antigens are in parallel with the QFT-IT assay and recent re- specific for MTB complex and few other my- sults have indeed demonstrated the usefulness of cobacteria such as M. kansasii and M. szulgai the HBHA-based assay to discriminate, among and are not found in atypical mycobacteria and in subjects infected with M. tuberculosis, those with the vaccine strain BCG, which lacks the region active TB from those with latent TB35. It would of difference 1 (RD1), encoding in MTB com- be of great interest to investigate whether in pa- plex for these antigens29. IGRAs measure the tient with TB spondylodiscitis the immune re- presence of effector T cells in the blood that are sponse against HBHA is similar to what seen in specific for M. tuberculosis. Effector T cells can patients with pulmonary TB, since this may open readily respond to an antigenic stimuli by secret- the possibility of using IGRA to guide in the di- ing cytokines, and are different from memory T agnosis of spondylodiscitis. cells that require more time (>24 hours). Effector In any case, the use of the IGRAs currently li- T cells are present only when the immune system censed provides more specific and sensitive re- is currently exposed to the antigenic stimuli, sults compared to TST for the diagnosis of TB which in this case is M. tuberculosis. A positive infection and shall be used to inquiry whether a IGRAs gives therefore an indication of an M. tu- TB spondylodiscitis is suspected. berculosis infection, but cannot distinguish, simi- In a recent report, the sensitivity of these larly to TST, between an active TB from a latent IGRAs in patient with Pott’s disease was found infection. The Quantiferon TB Gold In Tube at 84%, with a specificity of 95%, which was (QFT-IT) is an IGRAs that measure the amount higher than the 30% sensitivity measured for the of interferon-γ secreted following restimulation, combined smear plus culture classical microbio- and many laboratories provide to the clinicians logical tests24. Interestingly, when the immuno- the results in UI/ml. Many researchers have at- logical assay such as IGRA where combined tempted to correlate the amount of interferon-γ with the smear and culture, sensitivity reached secreted with TB status but so far unclear corre- 88%. These results, which certainly deserve to be lation has been found. confirmed in larger studies including more pa- The possibility to identify an immunological tients, underscore the usefulness of these im- correlate of TB disease would be of paramount munological assay in the diagnosis of spinal TB. importance in the case of non-pulmonary TB Moreover, it has been demonstrated the disap- such as spondylodiscites, where the possibility to pearance of the interferon-γ responses in patients link a clinical suspicion with an immunological who received effective therapy and persistent re- data strongly indicative of active TB would cer- sponse in patients who experienced failure of tainly help the implementation of an effective an- therapy36. Hence, IGRAs may also be useful in ti-TB therapy. To address this issue modified ver- monitoring TB therapy in patients with spinal sion of IGRAs have been developed, including TB. the use of different mycobacterial antigens to res- timulate T cells in vitro, the measurements of more and/or different cytokines such as TNF and IL-2 or chemokines such as IP-1030. So far, these Conclusions attempts have been only partly successful and more studies are required to identify immunolog- TB is still one of the most important infectious ical correlates of TB disease or latent TB. One of disease in the world, with a very high incidence the most promising evolution of these IGRAs in- in developing countries. The new migratory volves the use of another M. tuberculosis antigen, trends, with the arrival of workers from countries named the heparin binding haemagglutinin (HB- where TB is endemic to nations where TB was HA)14. It has been shown that T cells obtained thought to be conquered, is determining the

76 Microbiological and immunological diagnosis of tuberculous spondylodiscitis reemergence of TB disease, including less com- 6) DELOGU G, FADDA G. The quest for a new vaccine mon clinical manifestations such as Pott’s dis- against tuberculosis. J Infect Dev Ctries 2009; 3: 5-15. ease. Diagnosis of tuberculous spondylodiscitis 8) BRENNAN MJ, FRUTH U, MILSTIEN J, TIERNAN R, DE is a challenging task, and the clinician has to de- ANDRADE NS, CHOCARRO L. Development of new : a global perspective on fine a correct diagnostic protocol to maximize regulatory issues. PLoS Med 2007; 4: e252. the chances to identify the etiologic agent. The 9) BARRY CE, III, SLAYDEN RA, SAMPSON AE, LEE RE. introduction of the IGRAs for the immunological Use of genomics and combinatorial chemistry in diagnosis of TB could be of great usefulness and the development of new antimycobacterial drugs. shall be used before biopsy or surgery is per- Biochem Pharmacol 2000; 59: 221-231. formed. A definitive diagnosis requires the iden- 10) LE PAGE L, FEYDY A, RILLARDON L, DUFOUR V, L E HÉ- tification of the bacilli in the biological sample NANFF A, TUBACH F, B ELMATOUG N, ZARROUK V, GUIGUI P, F ANTIN B. Spinal tuberculosis: a longitudi- by at least one of the currently used methods nal study with clinical, laboratory, and imaging out- (microscopy, culture or molecular). When TB is comes. Semin Arthritis Rheum 2006; 36: 124-129. suspected, molecular diagnosis shall always be 11) COTTLE L, RIORDAN T. Infectious spondylodiscitis. J considered and can provide rapid and reliable re- Infect 2008; 56: 401-412. sults that could guide the adoption of proper ther- 12) KAUFMANN SH, COLE ST, MIZRAHI V, RUBIN E, apeutic regimes. NATHAN C. Mycobacterium tuberculosis and the host response. J Exp Med 2005; 201: 1693-1697. 13) RUSSELL DG. Who puts the tubercle in tuberculo- –––––––––––––––––––– sis? Nat Rev Microbiol 2007; 5: 39-47. Acknowledgements 14) MITTRUCKER HW, STEINHOFF U, KOHLER A, KRAUSE M, LAZAR D, MEX P, M IEKLEY D, KAUFMANN SH. This work has been funded by MIUR Prin 2008 Poor correlation between BCG vaccination-in- Y8RZTF. duced T cell responses and protection against tu- berculosis. Proc Natl Acad Sci USA 2007; 104: 12434-12499. 15) PETHE K, ALONSO S, BIET F, D ELOGU G, BRENNAN MJ, References LOCHT C, MENOZZI FD. The heparin-binding haemag- glutinin of M. tuberculosis is required for extrapul- monary dissemination. Nature 2001; 412: 190-194. 1) WORLD HEALTH ORGANIZATION. Global Tuberculosis Control. Surveillance, Planning, Financing, 2009. 16) DELOGU G, SANGUINETTI M, POSTERARO B, ROCCA S, ZANETTI S, FADDA G. The hbhA gene of Mycobac- 2) BROSCH R, GORDON SV, MARMIESSE M, BRODIN P, terium tuberculosis is specifically upregulated in BUCHRIESER C, EIGLMEIER K, GARNIER T, G UTIERREZ the lungs but not in the spleens of aerogenically C, HEWINSON G, KREMER K, PARSONS LM, PYM AS, infected mice. Infect Immun 2006; 74: 3006-3011. SAMPER S, VAN SOOLINGEN D, COLE ST. A new evo- lutionary scenario for the Mycobacterium tubercu- 17) BARRY CE, III, BOSHOFF HI, DARTOIS V, D ICK T, losis complex. Proc Natl Acad Sci USA 2002; 99: EHRT S, FLY N N J, SCHNAPPINGER D, WILKINSON RJ, 3684-3689. YOUNG D. The spectrum of : re- thinking the biology and intervention strategies. 3) COLE ST, BROSCH R, PARKHILL J, GARNIER T, Nat Rev Microbiol 2009; 7: 845-855. CHURCHER C, HARRIS D, GORDON SV, EIGLMEIER K, 18) BOSHOFF HI, BARRY CE, III. Tuberculo- GAS S, BARRY CE 3RD, TEKAIA F, B ADCOCK K, sis–metabolism and respiration in the absence of BASHAM D, BROWN D, CHILLINGWORTH T, C ONNOR R, growth. Nat Rev Microbiol 2005; 3: 70-80. DAVIES R, DEVLIN K, FELTWELL T, G ENTLES S, HAM- LIN N, HOLROYD S, HORNSBY T, J AGELS K, KROGH A, 19) LAWN SD, WOOD R, WILKINSON RJ. Changing con- MCLEAN J, MOULE S, MURPHY L, OLIVER K, OS- cepts of “latent tuberculosis infection“ in patients liv- BORNE J, QUAIL MA, RAJANDREAM MA, ROGERS J, ing with HIV infection. Clin Dev Immunol 2011; 2011. RUTTER S, SEEGER K, SKELTON J, SQUARES R, 20) HERNANDEZ-PANDO R, JEYANATHAN M, MENGISTU G, SQUARES S, SULSTON JE, TAYLOR K, WHITEHEAD S, AGUILAR D, OROZCO H, HARBOE M, ROOK GA, BJUNE BARRELL BG. Deciphering the biology of Mycobac- G. Persistence of DNA from Mycobacterium tuber- terium tuberculosis from the complete genome culosis in superficially normal lung tissue during sequence [see comments] [published erratum ap- latent infection. Lancet 2000; 356: 2133-2138. pears in Nature 1998; 396(6707): 190]. Nature 21) NEYROLLES O, HERNANDEZ-PANDO R, PIETRI-ROUXEL F, 1998; 11: 537-544. FORNES P, TAILLEUX L, BARRIOS PAYAN JA, PIVERT E, 4) COLE ST. Comparative and functional genomics of BORDAT Y, A GUILAR D, PRÉVOST MC, PETIT C, GICQUEL the Mycobacterium tuberculosis complex. Micro- B. Is adipose tissue a place for Mycobacterium tu- biology 2002; 148: 2928. berculosis persistence? PLoS ONE 2006; 1: e43. 5) KAUFMANN SH, MCMICHAEL AJ. Annulling a danger- 22) SHERMAN DR, GAGNEUX S. Estimating the mutation ous liaison: vaccination strategies against AIDS rate of Mycobacterium tuberculosis during infec- and tuberculosis. Nat Med 2005; 11: S33-44. tion. Nat Genet 2011; 43: 400-401.

77 G. Delogu, A. Zumbo, G. Fadda

23) FORD CB, LIN PL, CHASE MR, SHAH RR, IARTCHOUK Genome plasticity of BCG and impact on vaccine O, GALAGAN J, MOHAIDEEN N, IOERGER TR, SAC- efficacy. Proc Natl Acad Sci USA 2007; 104: CHETTINI JC, LIPSITCH M, FLYNN JL, FORTUNE SM. 5596-5601. Use of whole genome sequencing to estimate the 31) GOLETTI D, RAJA A, SYED AHAMED KB, RODRIGUES mutation rate of Mycobacterium tuberculosis dur- C, SODHA A, CARRARA S, VERNET G, LONGUET C, ing latent infection. Nat Genet 2011; 43: 482-486. IPPOLITO G, THANGARAJ S, LEPORTIER M, GIRARDI E, 24) RAFAILIDIS PI, AVRAMOPOULOS I, SAPKAS G, FALAGAS LAGRANGE PH. Is IP-10 an accurate marker for de- ME. Multidrug-resistant tuberculous spondy- tecting M. tuberculosis-specific response in HIV- lodiscitis: need for aggressive management and infected persons? PLoS ONE 2010; 5: e12577. drug susceptibility testing of Mycobacterium tu- 32) MASUNGI C, TEMMERMAN S, VAN VOOREN JP, berculosis isolates. J Infect 2006; 52: e35-e37. DROWART A, PETHE K, MENOZZI FD, LOCHT C, MAS- 25) KUMAR R, DAS RK, MAHAPATRA AK. Role of inter- CART F. Differential T and B cell responses against feron gamma release assay in the diagnosis of Mycobacterium tuberculosis heparin-binding Pott disease. J Neurosurg Spine 2010; 12: 462- hemagglutinin adhesin in infected healthy individ- 466. uals and patients with tuberculosis. J Infect Dis 26) DROUILLON V, D ELOGU G, DETTORI G, LAGRANGE 2002; 185: 513-520. PH, BENECCHI M, HOURIEZ F, B AROLI K, ARDITO F, 33) TEMMERMAN S, PETHE K, PARRA M, ALONSO S, TORELLI R, CHEZZI C, FADDA G, HERRMANN JL. Mul- ROUANET C, PICKETT T, D ROWART A, DEBRIE AS, ticenter evaluation of a transcription-reverse tran- DELOGU G, MENOZZI FD, SERGHERAERT C, BRENNAN scription concerted assay for rapid detection of MJ, MASCART F, L OCHT C. Methylation-dependent mycobacterium tuberculosis complex in clinical T cell immunity to Mycobacterium tuberculosis specimens. J Clin Microbiol 2009; 47: 3461-3465. heparin-binding hemagglutinin. Nat Med 2004; 27) PIERSIMONI C, BORNIGIA S, GHERARDI G. Perfor- 10: 935-941. mance of a commercial nucleic acid amplification 34) ZANETTI S, BUA A, DELOGU G, PUSCEDDU C, MURA test with extrapulmonary specimens for the diag- M, SABA F, P IRINA P, G ARZELLI C, VERTUCCIO C, nosis of tuberculosis. Eur J Clin Microbiol Infect SECHI LA, FADDA G. Patients with pulmonary tuber- Dis 2011 Jun 24. culosis develop a strong humoral response against 28) LECOUVET F, I RENGE L, VANDERCAM B, NZEUSSEU A, methylated heparin-binding hemagglutinin. Clin Di- HAMELS S, GALA JL. The etiologic diagnosis of in- agn Lab Immunol 2005; 12: 1135-1138. fectious is improved by amplification- 35) DELOGU G, BUA A, PUSCEDDU C, PARRA M, FADDA based DNA analysis. Arthritis Rheum 2004; 50: G, BRENNAN MJ, ZANETTI S. Expression and Purifi- 2985-2994. cation of Recombinant Methylated HBHA in My- 29) DIEL R, GOLETTI D, FERRARA G, BOTHAMLEY G, CIR- cobacterium smegmatis. FEMS Microbiol Lett ILLO D, KAMPMANN B, LANGE C, LOSI M, MARKOVA R, 2004; 239: 33-39. MIGLIORI GB, NIENHAUS A, RUHWALD M, WAGNER D, 36) DELOGU G, CHIACCHIO T, VANINI V, B UTERA O, CUZZI ZELLWEGER JP, HUITRIC E, SANDGREN A, MANISSERO G, BUA A, MOLICOTTI P, Z ANETTI S, LAURIA FN, D. Interferon-γ release assays for the diagnosis of GRISETTI S, MAGNAVITA N, FADDA G, GIRARDI E, GO- latent M. tuberculosis infection: A systematic re- LETTI D. Methylated HBHA produced in M. smeg- view and meta-analysis. Eur Respir J 2011; 37: matis Discriminates between Active and Non-Ac- 88-89. tive Tuberculosis Disease among RD1-Respon- 30) BROSCH R, GORDON SV, GARNIER T, E IGLMEIER K, ders. PLoS ONE 2011; 6: e1815. FRIGUI W, VALENTI P, D OS SANTOS S, DUTHOY S, 37) CARRARA S, VINCENTI D, PETROSILLO N, AMICOSANTE LACROIX C, GARCIA-PELAYO C, INWALD JK, GOLBY P, M, GIRARDI E, GOLETTI D. Use of a T cell-based GARCIA JN, HEWINSON RG, BEHR MA, QUAIL MA, assay for monitoring efficacy of antituberculosis CHURCHER C, BARRELL BG, PARKHILL J, COLE ST. therapy. Clin Infect Dis 2004; 38: 754-756.

78