Identification of mycobacteria species by molecular methods

Abstract

In this study, mycobacteria, which were previously identified as Mycobacterium tuberculosis complex (MTC), and mycobacteria other than tuberculosis (MOTT) with cord factor and the p‐nitro‐alpha‐acetyl‐amino‐beta‐hydroxypropiophenone (NAP) test were reanalysed using the polymerase chain reaction—restriction fragment length polymorphism (PCR‐RFLP) analysis method in order to confirm the identification, and at the same time, species accepted as MOTT were identified. Although the results of the NAP test were obtained within 3‐5 days, the PCR‐RFLP results were obtained in 1 day. Ten species identified as MTC with the NAP test and cord factor were confirmed with the PCR‐RFLP method. Fourteen species accepted as MOTT were identified as Mycobacterium species with the evaluation of the bands observed after the restriction of PCR product with the PCR‐RFLP method. These were as follows: three species Mycobacterium intracellulare type I, two species Mycobacterium phlei, two species Mycobacterium kansasii, one species Mycobacterium fortuitum type I, one species Mycobacterium gordonae type I, one species Mycobacterium abscessus type I, one species Mycobacterium scrofulaceum, one species Mycobacterium szulgai type I, one species Mycobacterium avium type II, and one species Mycobacterium terrae. Hence, the results of both the cord factor and the NAP test were confirmed with the molecular method, and at the same time, mycobacteria species identification was made by determining the fastest, easiest, and the most accurate result‐giving method. Because PCR‐RFLP is a very rapid method that provides exact identification of mycobacteria species, it can be performed in routine procedures.

1. INTRODUCTION

In the past two decades, there has been a dramatic increase in the numbers of diseases caused by Mycobacterium tuberculosis complex (MTC) and other non‐tuberculous mycobacteria. Therefore, there is great pressure on clinical laboratories to rapidly and accurately detect and identify clinically important mycobacteria.1 Differentiation of mycobacteria to species level is currently done by time‐consuming evaluation of phenotypic and biochemical characteristics.2 Additional techniques such as thin layer chromatography, gas‐liquid chromatography, high‐performance liquid chromatography, and gene sequencing are powerful tools but unfortunately remain limited to reference laboratories. Polymerase chain reaction—restriction fragment length polymorphism (PCR‐RFLP) analysis (PRA) of the hsp65 gene present in all mycobacteria offers an easy, rapid, and inexpensive procedure to identify several mycobacterial species in a single experiment.3 In this study, PCR‐RFLP and phenotypical identification were compared in a routine setting.

2. MATERIALS AND METHODS

All of the patient specimens were processed by the N‐acetyl‐L‐cysteine‐NaOH decontamination procedure recommended by the BACTEC 460‐TB system manufacturer (Becton Dickinson). A total of 0.5 mL of processed specimen was used to inoculate a BACTEC 12B medium vial and Lowenstein Jensen (LJ) medium.4 BACTEC 12B bottles with growth index values of ≥50 were vortexed briefly. Several drops of broth were smeared and stained with ehrlich‐ziehl‐nielsen (EZN). Serpentine cording was defined as tight, rope‐like aggregates of acid‐fast bacilli in which the long axes of the bacteria lie parallel to the long axis of the cord.5 Samples from EZN positive bottles without evidence of bacterial contamination were inoculated into a p‐nitro‐alpha‐acetyl‐amino‐beta‐hydroxypropiophenone (NAP) test bottle.6 The BACTEC NAP test utilises NAP to inhibit the growth of mycobacteria belongings to the MTC while not inhibiting or only partially inhibiting the growth of mycobacteria other than the MTC.7

2.1. DNA extraction

Fresh colonies were collected from LJ medium, mixed with 200 μL distilled water, and suspension was obtained. Acid‐fast bacteria were heated at 80°C for 1 hour 15 minutes, and then, mycobacterial DNA was isolated from the sample using the Nucleospin tissue isolation kit (Macherey‐Nagel GmbH).8

2.2. PCR

A total of 5.0 μL of the DNA extract was added to a reaction tube containing 45 μL of the PCR mixture (500 mM KCl, 100 mM Tris‐HCl [pH 8.3], 1.5 mM MgCl₂, 1 μL [each] deoxynucleoside triphosphate [Promega], 0.5 μM [each] primers TB11 [5′‐ACC AAC GAT GGT GTG TCC AT‐3′] and primers TB12 [5′‐CTT GTC GAA CCG CAT ACC CT‐3′] [Promega], and 0.25 μL of Taq polymerase [Promega]). The amplification was carried out using SANYO MIR D 40 thermal cycler. The reaction tube was first heated for 10 minutes at 95°C and then subjected to 35 cycles of amplification (1 minute at 95°C, 1 minute at 55°C, 2 minutes at 72°C), followed by 10 minutes extension at 72°C.2, 4, 8 The presence of amplified product was confirmed by agarose gel electrophoresis.

2.3. Restriction analysis

For BstEII digestion, 14 μL PCR product was added directly to a mixture containing 1.5 μL enzyme, 3 μL of restriction buffer R⁺, and 1.5 μL of water, and the mixture was incubated 24 hours at 37°C. Similarly, 14 μL of product was digested at 37°C in a solution containing HaeIII enzyme, the corresponding buffer O⁺, and water.2, 4, 8

2.4. Evaluation of restriction patterns

After digestion, 2 μL of gel loading buffer (10× loading buffer: 500 mg bromophenol blue, 5 mL glycerol, and 5 mL distilled water) was added, and 8 μL of the mixture was loaded onto a NuSieve 3:1 agarose gel (FMC Bioproducts). Fragments were visualised by ethidium bromide staining and UV light.2, 4, 8 RFLP patterns were compared by using the PCR‐RFLP algorithm (from http://www.hospvd.ch:8005).

3. RESULTS

Twenty‐four mycobacterium species, which were accepted as MTC and mycobacteria other than tuberculosis (MOTT) and which were obtained from the examination material of a case suspected with tuberculosis attending the Microbiology and Clinical Microbiology Department of the Trakya University Medical Faculty, were included in the study.

In the first instance, whether or not these species formed the cord factor was investigated and it was found that 10 out of the 24 samples were cord factor positive (+) and these were accepted to be M. tuberculosis. The other 14 species which were cord factor negative (−) were accepted to be MOTT.

In the second stage, the NAP test was applied to these species. Ten species which inhibited NAP were accepted as MTC and 14 species which did not inhibit NAP were accepted as MOTT.

In the third stage, the PRA method was applied to these species. Species identification was made by restriction of PCR products belonging to the heat shock protein region of all mycobacteria of 65 kDa, using restriction enzymes.

Ten species, which were identified as the MTC with the cord factor and the NAP test, were accepted to be MTC with the PCR‐RFLP method. However, 14 species accepted to be MOTT were identified with the PCR‐RFLP method. These were as follows: 3 species of Mycobacterium intracellulare type I, 2 species of Mycobacterium phlei, 2 species of Mycobacterium kansasii, 1 species of Mycobacterium fortuitum type I, 1 species of Mycobacterium gordonae type I, 1 species of Mycobacterium abscessus type I, 1 species of Mycobacterium scrofulaceum, 1 species of Mycobacterium szulgai type I, 1 species of Mycobacterium avium type II, and 1 species of Mycobacterium terrae.

The band pattern numbers and the names of 24 species restricted with BstEII and HaeIII restriction enzymes are shown in Table 1.

Table 1.

Identification of 24 species

	BstEII	HaeIII	Species		BstEII	HaeIII	Species
1	240.210	140.80.60	Mycobacterium phlei	13	240.120.100	145.130.60	Mycobacterium intracellulare tip I
2	240.120.85	150.130.70	Mycobacterium tuberculosis	14	325.130	185.140	Mycobacterium terrae
3	240.120.85	150.130.70	M. tuberculosis	15	240.210	130.105.80	Mycobacterium kansasii
4	240.120.85	150.130.70	M. tuberculosis	16	240.120.85	150.130.70	M. tuberculosis
5	240.120.85	150.130.70	M. tuberculosis	17	240.120.85	145.120.55	Mycobacterium fortuitum tip I
6	240.120.85	150.130.70	M. tuberculosis	18	240.120.85	160.115.60	Mycobacterium gordonae tip I
7	240.120.85	150.130.70	M. tuberculosis	19	240.120.100	145.130.60	M. intracellulare tip I
8	240.120.85	150.130.70	M. tuberculosis	20	240.210	145.70.60	Mycobacterium abscessus tip I
9	240.120.85	150.130.70	M. tuberculosis	21	440	130.105.70	Mycobacterium szulgai tip I
10	240.120.85	150.130.85	M. tuberculosis	22	240.210	145.130.95	Mycobacterium scrofulaceum
11	240.210	140.90.60	M. phlei	23	240.210	130.105.80	M. kansasii
12	240.120.100	145.130.60	M. intracellulare tip I	24	240.210	130.105.0	Mycobacterium avium tip II

Band pattern figures maintained by cutting the amplifıcation product of gene zone coding Head shock protein by HaeIII and bastell enzymes are shown in Figures 1, 2, 3, 4, 5, 6.

Figure 1.

HaeIII

Figure 2.

BstEII

Figure 3.

BstEII

Figure 4.

HaeIII

Figure 5.

BstEII

Figure 6.

HaeIII

4. DISCUSSION

Twenty‐four species of mycobacteria were included in the study. Ten species, which were identified as the MTC with the cord factor and the NAP test, were accepted to be MTC with the molecular method. Fourteen species accepted as MOTT, however, were identified to be different mycobacteria types with the molecular method we have used. Beginning from 1993, Tenover et al9 have underlined the importance of reorganising tuberculosis laboratories in order to obtain rapid diagnosis and they have targeted their statement to the facts that EZN diffusion results are obtained within 24 hours, the growth and identification of M. tuberculosis strains are completed within 10‐14 days, and the resistance experiment results are given in 15‐30 days. In a study by Koksalan et al, the specificity, sensitivity, and the positive and negative predictive values of the cord factor in the EZN diffusion of 77 positive BACTEC cultures were found to be 92.6%, 100%, 100%, and 64.3%, respectively. They have stated that, due to the low negative predictive value, in our country, in situations where the cord factor is found positive, a pre‐diagnosis of MTC growth is made, but in situations where the cord factor is found to be negative, a pre‐diagnosis based only on the cord factor should not be made and they have also stated that another identification method such as hsp65 PRA should be used at the same time.10 Badak et al in İzmir, in a study evaluating the cord formation with Kinyoun dye of 370 positive MB/Bac cultures, have found specificity, sensitivity, and positive and negative predictive values to be 88.2%, 97.4%, 99.2%, and 69.7%, respectively, by a first observer and 90.6%, 52.3%, 82.8%, and 69.7%, respectively, by a second observer. As a result, it has been underlined that the observation of cord formation in diffusions by experienced microbiologists in laboratories where the growth rate of MTC is high, is rapid and reliable.11

While cord formation and BACTEC NAP tests have been used as the golden standard until the beginning of the 1990s in the identification of mycobacteria types, it has left its place to molecular methods in recent years.10 Although biochemical tests are very cheap and enable identification to species level, it is not recommended in routine tuberculosis laboratories in our country, because it requires a great deal of effort.8

Studies concerning the specificity, quickness, cost effectiveness, and the productivity of the PCR‐RFLP analysis method of the hsp65 gene that was on the agenda in the beginning of 1993, were reported in a number of publications, and this made it possible to be routinely used.2, 3, 12

da Silva Rocha et al stated that the PCR‐RFLP method could be helpful in the investigation of unidentified mycobacterial samples and was an economic and easy method for all the mycobacterial isolate differentiations they had studied.13 de Magalhaes et al compared the reliability of the hsp65 RFLP method with conventional culture and biochemical test results, for the identification of mycobacteria species obtained from 247 examination materials and 88 positive cultures. They have reported that this method was effective even for bacterial cultures with very low growth index and reduced the time for diagnosis but stated that when applied to clinical samples showed low sensitivity.14

The PCR‐RFLP analysis of the hsp65 gene was tried with the DNA obtained from EZN positive sputum, but due to the presence of materials inhibiting restriction enzyme, it was found to be unsuccessful. MTC cannot be differentiated from M. tuberculosis with this method. This situation is due to the group's advanced level of genetic interactions, and this problem is also valid for other identification methods based on the PCR method.2 Wong et al have reported the sensitivity and the specificity of the PCR‐RFLP analysis method of the hsp65 gene for M. tuberculosis and M. avium to be 100%. The results were encouraging even though the number of isolates tested were low, and when compared with other tests, it was reported to be cost‐effective (1.50 US $ for each sample) and the use in routine clinical laboratories to be rather feasible.1

The Tb11 and Tb12 primers used in the PCR‐RFLP analysis of the hsp65 gene can propagate the 65 kDa heat shock protein region of bacteria which are not mycobacterium (Nocardia brasiliensis, Streptococcus albus, Rhodococcus equi, and so on). However, because the EZN staining of the culture is performed usually before the PCR‐RFLP analysis and because the restriction patterns formed are not going to be mixed, this is not a problem.4 The PCR‐RFLP analysis method of the hsp65 gene is a standard method which gives results in 1 day and can identify mycobacteria to species level without the need for hybridation, probe specific to the species and radioactivity.2, 3, 4, 15

The cord formation and BACTEC NAP test are partial identification methods, which can differentiate MTC species from MOTT species, but they cannot differentiate MOTT types to species level. The PCR‐RFLP analysis method of hsp65 gene is an identification method that can differentiate MOTT types from the species level. The cord formation and NAP test were found to be 100% in agreement with the molecular method used in this study with the MTC and types accepted as MOTT. The important reason behind the proposal of the use of this method is that using this molecular method, the species in this study were identified in 1 day and is a rather economic method.

Kunduracılar H. Identification of mycobacteria species by molecular methods. Int Wound J. 2020;17:245–250. 10.1111/iwj.13238