Clin Microbiol Infect. 2013 Mar;19(3):273-8.

Evaluation of the detection of Mycobacterium tuberculosis with metabolic activity in culture-negative human clinical samples.

Cubero N, Esteban J, Palenque E, Rosell A, Garcia MJ.

Departamento de Medicina Preventiva, Facultad de Medicina, Universidad Autonoma de Madrid, Madrid, Spain.



Mycobacterium tuberculosis is assumed to remain in a quiescent state during latent infection, being unable to grow in culture. The aim of this study was to evaluate the detection of viable but non-cultivable bacilli with metabolic activity in human clinical samples using a procedure that is independent of the immunological status of the patient. The study was performed on 66 human clinical samples, from patients subjected to routine diagnosis to rule out a mycobacterial infection. Specimens from pulmonary and extra-pulmonary origins were verified to contain human DNA before testing for M. tuberculosis DNA, rRNA and transient RNA by real-time quantitative PCR.

Clinical records of 55 patients were also reviewed. We were able to detect viable but non-cultivable bacilli with a metabolic activity in both pulmonary and extra-pulmonary samples. Mycobacterium tuberculosis RNA was detected in the majority of culture-positive samples whereas it was detected in one-third of culture-negative samples, 20% of them showed metabolic activity. Amplifications of the ftsZ gene and particularly of the main promoter of the ribosomal operon rrnA, namely PCL1, seem to be good targets to detect active bacilli putatively involved in latent infection. Moreover, this last target would provide information on the basal metabolic activity of the bacilli detected.

PMID: 22360423


Supplementary Information:

Tuberculosis (TB) is a disease with complex pathogenic mechanisms, the lung being the main tissue targeted by the infection. The bacilli, once inside the human host, can cause either a latent disease (about 90% of cases) or an active disease (about 10% of cases), this last a condition that may also result from the progression of the endogenous reactivation of a latent infection.

Latent TB infection (LTBI) is a host’s condition characterized by the presence of a persistent immune response, with absence of symptoms or signs of the disease. In this sense, the clinical definition of LTBI appears clear, however the state of the bacteria in such a condition is rather poorly characterized. Traditionally, latent TB is supposed to be caused by viable but non-cultivable bacilli (live bacteria that are unable to form colonies in vitro) that remain quiescent in the lung granulomas. In this context, the metabolic state of the M. tuberculosis during latent infection remains to be characterized and, to date, it is not possible to determine if the bacilli are actually alive or dead.

Experimental approaches have been developed to get insights into the situation of the latent bacilli. A range of models, from in vitro to animal models are currently being applied, however, although of interest,  none of them seems to answer all the questions opened to understand the status of the tubercle bacilli during latency.

New controversial data indicate that M. tuberculosis DNA can be detected in tissues without granulomas, and the recent dynamic hypothesis of LTBI suggests that the quiescent bacilli are active and may be spread out from the granulomas during latency. This is in accordance with the recent consideration of M. tuberculosis infection as a continuous spectrum expanding from sterilizing immunity to clinical disease. Therefore, within the host, the bacilli could be detected under several conditions from dividing to non-dividing.

Molecular amplification methods are considered the more reliable procedures to study and characterize infectious agents. Diagnostic assays based on nucleic acid amplification methods, including PCR, are currently used for a rapid TB diagnosis. However, their usefulness for LTBI diagnosis remains uncertain. In fact, from a general perspective, nucleic acid amplifications are considered useless for the diagnosis of LTBI.

The purpose of this study was to evaluate the detection of viable but non-cultivable bacilli during infection in human host. We studied the presence of M. tuberculosis with metabolic activity in culture-negative samples from those received for routine diagnosis from two different hospitals. No selection criteria of the samples were established, allowing the collection of pulmonary and extra-pulmonary specimens.

We aimed to detect M. tuberculosis with metabolic activity through detection of bacterial transient RNA in clinical samples that were negative for this bacteria by using standard microbiological diagnostic procedures, namely: acid-fast stain, in vitro culture and standard commercial probes.

DNA is a stable molecule with a discrete time of life as free molecule in a clinical sample before degradation. Due to this characteristic the detection of M. tuberculosis DNA in culture-negative samples is usually explained by the presence of dead or non-viable bacilli. On the contrary, the RNA is quite unstable in a sample, due to the high activity of RNAses. Therefore, the detection of cDNA (the complementary copy of the RNA) is usually considered as a confirmation of the presence of viable bacilli.

On these bases we would like to go further, and to determine if the main metabolic activity of VBNC bacilli could be detected. To do so, we search for the detection of synthesis of ribosomal RNA, because ribosomes are pivotal in all the global metabolic activity of any bacteria.

To obtain information about the basic metabolism of the bacilli we aimed to detect the active synthesis of the ribosomal RNA that reflects the basal metabolic activity of the cells. Ribosomal RNA is synthesized as a larger molecule, the pre-rRNA, which is then digested by RNAses to give a shorter mature molecule able to link to ribosomal proteins and to form ribosomes.

It is known that the stability of the pre-rRNA is similar to that of other mRNA in the cell, being that the mature 16S rRNA is more stable probably because it is linked to bacterial ribosomal proteins as part of the ribosomes.

This means that the detection of 16S rRNA may indicate that the bacteria were alive, but their metabolic activity could not be confirmed at this step. For this reason, we amplified the pre-rRNA product derived from the main ribosomal RNA promoter of M. tuberculosis, namely PCL1, as well as the mature product, the 16S rRNA. We performed quantitative reverse transcription PCR (qRT-PCR), this technique is useful to detect and quantify the presence of bacterial RNA in clinical samples.

We detected the presence of the 16S rRNA in 25 samples, including 18 culture-negative samples. However, the detection of other transcripts involved in the bacterial growth (such as ftsZ and PCL1) was more frequent in culture-positive samples. This is expected because of the physiological role of the respective gene products. In seven of the culture negative samples that only amplified the 16S rRNA, IS6110 was not detected so they may have contained non-tuberculous mycobacteria.

According to our results, bacilli putatively involved in LTBI were detected in 30.2% of our culture-negative samples and bacilli with metabolic activity were identified in 20.8% of them from both pulmonary and extra-pulmonary origins. These data are in agreement with the global standard level of LTBI. The detection of mycobacterial cDNA in pulmonary and extra-pulmonary culture-negative samples agreed with previous data suggesting that M. tuberculosis may establish a latent infection in both types of anatomical locations with or without granulomas. Our procedure allows an identification of LTBI that was independent of the immunological status of the patient. This is particularly relevant in the pediatric population.

Regarding the clinical impact of the qRT-PCR in TB diagnosis, the clinical records were available for 55 patients. All patients with culture-positive samples showed clinical data related to TB. Most of the patients with culture-negative samples (30/47) had clinical data that might be related to a mycobacterial infection. This was expected because the samples were collected from those received in the laboratories usually following a clinical suspicion of mycobacterial infection.

The main limitation of our study was the number of samples available. However, this limitation may reinforce the hypothesis that the number of samples containing viable but non-cultivable bacilli may be increased in a controlled study. Another limitation was that the level of DNA contaminating the cDNAs was unknown, and could only be approximated through the level of IS6110 detected (a multi-copy insertion sequence). We have no clear solution to solve this problem because it is recommended to avoid DNAse treatment to minimize the loss of this scarce material during further purifications.

Finally even though this is a rare event except for some South Asian countries, the possibility of viable but non-cultivable M. tuberculosis bacilli with a low or zero copy number of IS6110 in their genomes should also be considered. Other IS multi-copy elements could be targeted instead for DNA detection when these circumstances are suspected.

According to our results, we suggest that the PCL1 promoter of the rrnA operon of M. tuberculosis may be a target of interest to detect viable and metabolically active bacilli in culture-negative samples.

The advantages of this detection procedure to test for LTBI would be to avoid the secondary effects that may be observed with tuberculin skin test. Moreover, it would not be influenced by the immune system state of the patient as observed with that test and interferon-g release assays, it can be used advantageously in children and it can be performed on every type of sample, of either pulmonary or extra-pulmonary origin.

We conclude that detection of PCL1 is a good biomarker of M. tuberculosis to determine basic metabolic activity of viable bacteria. This biomarker has two main applications:

  1. Identification of latent infection, by using a procedure independent of the immunological status of the patient (this last the usual way to identify latent infection)
  2. Detection of bacilli with basic metabolic activity in the presence of drugs. The use of this marker will allow the differentiation between bacterial persistence (biomarker positive) and bacterial drug susceptibility (biomarker negative) thus helping in the accuracy of the drug’s activity in vitro.
Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann