Evaluation of FLASH - PCR forrapid detection of Mycobacterium tuberculosis from clinical specimens.

BACKGROUND AND OBJECTIVES: Tuberculosis (TB) is the oldest known bacterial disease in humans. Due to the rise of morbidity in recent years, early diagnosis of the disease is necessary.
MATERIALS AND METHODS: In this study we used Fluorescent Amplification-Based Specific Hybridization (FLASH) PCR to targetIS6110 for rapid detection of M. tuberculosis (MTB). To investigate the important factors influencing the risk of TB, data from patients and their medical records were analyzed. RESULT: The sensitivity and specificity of FLASH-PCR for detecting MTB were determined as 93.33% and 92.5%, respectively. The findings of this study have suggested that removal of the contaminants in FLASH-PCR sign ificantly reduced the detection time, and MTB was much more rapidly detected in the clinical specimens compared to the conventional culture and smear examination. Results of the medical survey showed that the majority of TB patients were males, over 51 years old, smokers, with pulmonary TB and normal chest X-ray (CXR).
CONCLUSION: MTB can be rapidly detected inclinical specimens using FLASH-PCR in comparison with culture and smear examination.

INTRODUCTION

Mycobacterium tuberculosis(MTB) is one of the important causes of millions ofdeaths among adults annually (1). Recently, in many regions of the world, increase in multi drug resistant tuberculosis (MDR-TB) has become an emerging problem for public health (1, 2). Rapid detection of MDR-TB is necessary andcruicial for treatment management (3, 4). MDR-TB causes higher mortality rates than drug susceptible tuberculosis and is even more important in patients with immunodeficiency (2, 5). Moreover, early diagnosis of MTB infection is essential tothe strategy selection for control of the spread and effective treatment (6, 7). Staining and direct microscopy of the pulmonary and extra pulmonary specimens lack sufficient sensitivity and specificity. The conventional techniques of MTB diagnosis are unreliable and time consuming (8, 9). Researchers are interested in devising rapid tests for detection of MTB (10, 9, 11). In recent years, themain focus has been on the development of PCR techniques (6) and many primer sets have been developed for specific detection ofMTB genes (11, 12).Fuorescent amplification-based specific hybridization PCR (FLASH-PCR) with the power of fluorescence detection after the amplification has been introduced as a rapid and sensitive method for identification of pathogenic bacteria (12, 13). Abramova et al. applied FLASH–PCR for specific identification of wheat pathogenic fungi S. tritici and S. nodorum(14). Ryazantsev et al. diagnosed toxigenic fungi of the genus Fusarium with the same PCR technique (12). Moreover, Ryazantsev and Zavriev have developed an efficient diagnostic method for the major potato viral pathogens based on FLASH–PCR(15). It should be noted that FLASH–PCR requires no expensive equipment and the results are obtained within a few hours (12).On the other hand, this method decreases the risk of the bench top contamination (15),whereas contamination is inevitable indetection by electrophoresis (14).

The repetitive insertion sequence IS6110 has been one of the excellent targets for PCR amplification (16-18) and DNA probes for diagnosis of MTB and epidemiological investigations since early 1990s (19-21). Multiplecopies of IS6110 (10 to 20 copies) exist in MTB (6, 22), whereas strains of M. bovis BCG contain only a single copy (16, 22). Therefore, targeting IS6110 is helpful to increasethe sensitivity of PCR over that obtained in the amplification of single-copygenes.

In this study, we aimed to develop a FLASH-PCR system as a rapid and sensitive method for the detection and identificationof MTB. The results obtained by FLASH-PCR technique was compared with those of culture and smear microscopy. Also, the important factors influencing the risk of TB disease such as age, gender, history of contact with TB patient, undelying diseases, and smokingwere investigated in the patients.

MATERIALS AND METHODS

Sample collection and bacterial isolation

During April 2009 until March 2010, 255 pulmonary and extra-pulmonary specimens were collected from patients suspected with TB hospitalized in at eaching hospital in North of Tehran.Clincal specimens included sputum, bronchoalveolar lavage (BAL), pleural fluids, pulmonary aspiration, tracheal aspirates, ascites, wound samples, urine, gastric fluids, synovial fluids, throat swabs, and abscesses. The specimens were tested directly by Ziehl–Neelsen staining and cultured on Lowenstein-Jensen (LJ) medium as the standard methods (17, 24).

DNA extraction with Sorbent method

Mucolysin solution was used in the purification step to prevent the inhibitory effects of mucosal materials on PCR. Qiagen kit was used for DNA extraction as instructed by the supplier (Qiagen, Germany). DNA was eluted in 50μl NSP-5 (elution buffer), intensely vortexed, and incubated at 55°C for 15 min. Finally, 30μl of supernatant containing DNA was used for PCR reaction.

DNA amplification

DNA amplification of the 123 bp IS6110 insertion element was performed with above mention primers(25). Briefly, PCR reaction was designed with 25 μl total reaction mixture containing 10 μl of 2X master mix reaction buffer (1 mM of each dNTP,fluorescent probes, and 2 mM MgCl2; Takarra, Japan), 1 μL of Taq polymerase (Roche, Germany), 5μL(500 ng) product of DNA extraction (DNA sample), 5 μL DDW, 1 μL forward and reverse primers at final concentration of 10 pmol and 3 μL of internal control (IC). The primers and the probe were used from a kit (DNA Technology Russia). The GeneAmp PCR system 9600 (PerkinElmer, Wellesley, MA, USA) was used for DNA amplification. The amplification cycles contained: 1 cycle,3min at 94°C for initial denaturation, 5 cycles, each of 1min at 94°C, and 1min at 67°C for denaturation and primer annealing, 45 cycles each of 30s at 94°C, 50s at 67°C for extension. PCR product tubes were transferred to Fluorescent Detector FD-12 (DNA Technology, Russia) and each tube was read and analyzed with GeneXpert software.

Survey of important factors influencing the risk of TB disease

A questionnaire was developed to collect social and demographic information of all TB cases (pulmonary and extra-pulmonary) such as age, gender, history of contact with TB patient, smoking and narcotics. Furthermore, medical documents of these patients were studied that contained a range of illnesses including bronchiectasis, cancer, sarcoidosis, pneumonia, fibrosis, pulmonary embolism, hydatid cyst, asthma, chronic obstructive pulmonary disease (COPD), chemical injury, symptoms of TB, normal or abnormal CXR, PPD skin test, and old or new TB (26).

Statistical Analysis

Data were processed and analyzed using statistical software such as Microsoft Excel 2010 and SPSS version 16.0. Comparisons between groups were made using chi square and Fisher’s exact test.

RESULTS

From AprIL 2009 until MarCH 2010, 255 TB suspected cases were examined, of which 30 subjects were positive in MTB FLASH-PCR assay.Most of the positive specimens were BAL samples (25 out of 30, 83%). The patients ranging from 1 to 97 years of age (average 56 years) were studied and most of them were 50 years old (164 out of 255, 64%). The age breakdown of 30 positive cases were as follows: 5 cases ≤ 30 (16.6%), 1 case 31-40 age (3.3%), 5 cases 41-50 (16.6%) and 19 cases ≤ 51 (63.5%). 106 out of total 255 subjects were females (41.6%) and 149 males (58.4%); in 30 positive cases, 9 cases were females (30%) and 21 males (70%).

Comparison between FLASH-PCR and microscopic and culture examination

FLASH-PCR was performed with IS6110 as specific target gene for MTB detection. For detection of FLASH-PCR products, PCR product tubes were transferred to Fluorescent Detector FD-12. The PCR reaction was validated by detection of fluorescent IC. In positive samples (MTB patients) fluorescence of MTB probe was detected (Fig. 1). Therefore, the products are easily identified without the extra step of electrophoresis and the risk of contamination.

Fig. 1

FLASH-PCR; upper fig: chart of flash PCR that detected with GeneXpert software.

Lower fig: gel electrophoresis; M: 100bp ladder, 2 &7 were negative sample, 1 was low positive sample.

Maximum positive detection was obtained by FLASH-PCR in 30 cases (11.7% of total subjects), followed by culture (15 cases, 5.8%) and smear examination (11 cases, 4.3%) (Fig. 2).

Fig. 2

Frequency of the positive samples (FLASH-PCR, culture, and smear examination).

Comparison of PCR results with both culture and microscopy examination showed that FLASH-PCR technique is more sensitive than other tests (Fig. 3) (p < 0.05), but specificity of this type of PCR was less than the one in gold standard test (culture) (Fig., p> 0.05).

Fig. 3

Comparison of FLASH-PCR with: a. Culture and b. Smear.

Documentation of medical records

In the second part, we studied the factors that might play an important role in the incidence of TB disease (Fig. 4). According to the laboratory results, among the 30 TB patients, 28 (93.3%) were pulmonary and 2 (6.66%) were extra-pulmonary cases (1 case urinary tract TB, and 1 case pleural TB).

Fig. 4

a. Symptoms of TB disease in the patients, b. Background of diseases of patients.

Six out of 255 subjects had a history of contact with TB patients in their familiesor relatives of whom only one case was reported positive for clinical tuberculosis (0.39%).Among 249 cases that had no previous contact with a TB patient, 29 (1.13%) were positive for TB.

Since the initial diagnosis of pulmonary tuberculosis is often based on abnormal findings in the CXR images of patients plus respiratory signs and symptoms, the key to the clinicaldiagnosis of tuberculosis is often under suspecion. In this study, 22 out of 255 cases had tuberculosis specific report of CXR with fibrotic lesions. 2 cases (0.78%) were positive for TB, 140 of 255 cases had non specific report of CXR for tuberculosis,26 of whom (1.01%) were TB positive. 93 of 255 cases had a normal CXR, 2 cases (0.78%) were TB positive.

Our results of 30 TB positive patients, revealed that 3 cases (10%) had positive PPD, and PPD was not reported in many patients.In total,16 subjects (7.1%) had positive PPD test among 225 cases that were TB negative.

Furthermore, in the present study, 10 common symptoms of tuberculosis were selectedincluding: fever, chills, sweating, night sweating, weight loss, dry cough, cough with secretions, coughing up blood, asthma, and chest pain (Fig.3a). Most of people with TB symptoms were asthmatic (225 cases, 83.3%), and showed symptoms such as weight loss (18 cases, 60%), fever (17 cases, 56.7%), dry cough (17 cases, 56.7%), and cough with secretions (15 cases, 50%). In addition, 11 common lung diseases of the patients were studied in their medical documents. (Fig. 3b). Among 30 TB positive cases, 4 cases suffered from bronchiectasis (13.3%), 4 cases had lung cancer (13.3%), 1 case had co-infection with sarcoidos is (3.3%), 11 cases with pneumonia (36.7%), 5 cases with chemical injury (16.7%), 1case with hydatid cyst (3.3%), 1 case with asthma (3.3%), and 8 cases with COPD (26.7%). Considering the known effects of smoking, narcotics and alcohol on the lung diseases, we investigated the relationship between these factors and lung diseases. Out of the 255 cases studied, 193 were non-smokers, 41 smokers, 8 smokers and consumer of narcotics, 3 smokers and consumer of alcohol, and 10 consumers of narcotics. Among 30 TB positive patients, 21 were nonsmokers (70%), 7 smokers (23.3%),and 2 smokers and narcotic consumers (6.7%).

DISCUSSION

The World Health Organization (WHO) declared TB asone of the the greatest health problems for the global public health (27, 28). Since early detection of MTB is an important emergency procedure for the control of tuberculosis (16),many researchers from different laboratories have used the PCR method for early diagnosis of MTB. In several studies, sensitivity of PCR with different DNA extraction methods has been reported between 80-90% for MTB complex diagnosis. The discrepancy in the findings could be related to the different type of PCR and primer sets. Various molecular methods have been applied to detect M. tuberculosis and gene mutations. The molecular methods such as PCR single stranded conformation polymorphism (PCR SSCP) assay, multiplex specific PCR assay, PCR RFLP analysis, PCR based dot blot hybridization strategy, molecular beacon assay, and direct sequencing assay (29-34) are each capable of detecting the bacterium, however they are time consuming, expensive and labor intensive. Real time PCR and GeneXpert MTB/RIF assays are highly sensitive and specific for the detection of M. tuberculosis and rifampicin resistance (35, 36). Importantly, one of the features of the MTB/RIF assay is its potential utility in resource poor settings. In such settings, there may be limited, if any, access to confirmatory phenotypic or genotypic drug susceptibility testing. In real time PCR technique the probe is used for detection and quantitative analysis. However, quantitative analysis is not as important as detection of MTB. In addition, real time PCR and MTB GeneXpert methods are dependant on specific real time PCR teromcycler. Therefore, an accurate, cost effective, rapid and easy-to-perform approach for identifying mycobacterial species is necessary.In recent studies FLASH-PCR technique has been used to detect various pathogens rapidly and specifically. Abramova et al.(14), used FLASH-PCR to detect the phytopathogens Septoria tritici and Stagonospora nodorum. Ryazantsev et al. (12)optimized FLASH-PCR for diagnosis of the toxigenic fungi F. sporotrichioides and obtained similar to those gained by the electrophoretic analysis. On the other hand, their results were similar to Ryazantsev and Zavriev’s studyfor diagnosis of potato viral pathogens (15), and detection of Septoria tritici in wheat seeds by Consolo et al.(10). In this study FLASH-PCR concomitant with culture and smear examination were carried out for rapid detection of MTB in clinical specimens. Compared with the culture and smear results, sensitivity and specificity of the FLASH-PCR method for MTB detection was 93.33% and 92.5%, respectively.This demonstrates that FLASH-PCR test has a higher sensitivity than conventional examination and can be performed in less than 2 days (17). In our study, the primer specific from IS6110 fragment was used in the FLASH-PCR for MTB detection, because there are multiple copies of this sequence in the genome of MTB which increases the accuracy, sensitivity, and specificity of this technique (23). Some culture-negative samples turned positive in PCR (18 samples). The loss of mycobacteria in cultures may be due to the decontamination process or structural damage to the bacteria or the death caused by drug therapy. It has been shown in previous studies that the bacterial count significantly reduced if the bacterial culture has not been performed immediately. Therefore, it is recommended that laboratories that perform MTB culture, must pay attention to the “time” factor (37). Exceptionally in the present study we gained negative PCR for a culture positive sample, which actually was a false negative result. This may be related to the low number of copies of IS6110 in the genome of the samples or non uniform distribution of bacteria in the sputum samples.

In this study, there was no significant difference between the number of male (58.4%) and female (41.6%) participants. These findings are consistent with previous studies (26), that reported a higher frequency of pulmonary tuberculosis in men, which is due to job and smoking and narcoticuse in male cases.In general, the diagnosis of pulmonary TB is performed with identification of acid fast bacilli from bronchial secretions, chest radiography and evidence of patient’s physical signs (38). The advantage of this study is that majority of patients had non specific CXR report for tuberculosis or had normal CXR (39, 40). Normal CXR and a negative smear is a challenge for the early diagnosis of pulmonary TB, and we recommend FLASH-PCR as a more sensitive diagnostic tool to avoid the delay in the specific diagnosis of tuberculosis. In the present survey, most of the patients were smokers and some of them were drinking alcohol or consumed narcotics. In several studies a relationship between smoking and TB has been shown and was reported that the prevalence of smoking is higher in males than in females. Thus, smoking has been known as a risk factor for pulmonary tuberculos is because of various deleterious effects on the lungs (especially in the upper zones of the lung) (41). Smoking and alcohol consumption (53.6% and 36.5%, respectively) were the important risk factors for the development of tuberculosis (3). Interestingly, 5 out of 30 TB positive patients (16.7%) were chemically injured. This finding implies that immune compromised patiens may be more prone to tuberclosis.

In conclusion, the findings of this study have suggested that MTB in clinical specimens can be rapidly detected using of FLASH-PCR associated with culture and smear examination. FLASH-PCR as rapid diagnosis can be used for HIV infected patients because of a higher susceptibility to MTB infection and so can be a great help for controlling MDR-TB. Moreover, FLASH-PCR technique allows results to be recorded without opening the test tubes and this means the removal of the contaminants in a conventional PCR, while contamination occurs during electrophoresis. Another advantage of this technique is the addition of specific primers including fluorescent probes that increased sensitivity and specificity of this technique.