Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) for rapid identification of Mycobacterium abscessus

Abstract

Background

Nontuberculous mycobacteria are increasingly being implicated in infections and have become an important cause of health care associated infections. Mycobacterium abscessus, a rapidly growing mycobacteria, is of particular concern as it tends to be resistant to commonly used therapeutic options. Conventional phenotypic methods for speciation of mycobacteria are time consuming, labor intensive and not always reliable. Molecular methods require expertise and are expensive. The study was used to evaluate the use of matrix associated laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) as a means of rapid identification of M. abscessus.

Methods

35 isolates of rapidly growing Mycobacterium from an outbreak of surgical site infections at a tertiary care hospital were identified using phenotypic methods. The mycobacterial isolates were inactivated and an extraction protocol was followed. These isolates were then analyzed by MALDI biotyper (Bruker Daltonics) using biotyper software 4.0 and the mycobacterial reference database v 2.0.

Results

All 35 isolates were identified as M. abscessus by MALDI biotyper but the scores obtained according to guidelines of the company were lower than previous studies with only 23 out of the 35 isolates having scores of greater 1.8 which was described as the minimum score to be achieved for reliable identification.

Conclusion

MALDI-TOF MS offers a rapid and inexpensive method for identification of Mycobacteria; however, the scores obtained in our study were lower than reported in other studies.

Introduction

Nontuberculous mycobacteria (NTMs) include all other mycobacterial species except the Mycobacterium tuberculosis complex and Mycobacterium leprae.¹ They have previously been referred to in the literature as “atypical mycobacteria” or “mycobacteria other than Mycobacterium tuberculosis (MOTT)”.¹ They are ubiquitous in the environment and can be colonizers on the skin, respiratory and gastrointestinal tracts of healthy individuals. Infections due to NTMs could be acquired by trauma, aerosol inhalation and occasionally nosocomially. Human to human transmission is rare. In 1954, Runyon classified them into four groups (I–IV) based on the growth rate, colony morphology and pigmentation. Slow growing NTM belong to Runyon group I–III and the rapid-growers belong to Runyon group IV.² Rapid growers are characterized by growth on solid media on subculture within 7 days.² The present study pertains to the detection of Mycobacterium abscessus classified in group IV of Runyon classification. Moore et al in 1953³ were the first to describe M. abscessus. This organism was isolated predominantly from patients of cholelithiasis and other intra-abdominal lesions who were subjected to laparoscopic intervention. M. abscessus was a common isolate from the surgical site infection subsequent to scope insertion.

The recognition of M. abscessus as an important human pathogen, capable of causing a wide spectrum of disease ranging from skin and soft tissue infections to disseminated infection, was possible after it was separated from the Mycobacterium chelonae which also belongs to same group IV.⁴

Routine species identification and differentiation of NTMs in the clinical laboratory is challenging as battery of biochemical tests are required which still may be unable to accurately identify less common species.⁵ Molecular methods using molecular probes for gene sequencing are technically demanding and relatively expensive for routine use.

Mass spectrometry appears to hold a prominent role in rapid and accurate identification of the NTMs. Recently, matrix-associated laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been used as a research tool in detection and identification of various bacterial isolates.

In brief the sample for analysis is mixed with an organic compound which is energy-absorbent called the matrix. On drying the matrix along with the sample crystallizes. A laser beam is used for ionization and desorption of the ions from the sample. This generates protonated ions, which are then accelerated at a fixed potential so as to separate them from each other on the basis of their mass-to-charge ratio (m/z).

In TOF analysis the m/z ratio is determined by the time taken for the ions to travel the length of a flight tube. On the basis of the information generated, a characteristic spectrum known as the Peptide mass fingerprint (PMF) is generated. Comparison of this PMF (of the unknown sample) with an available database enables the identification of the organism. A typical mass range of 2–20 kDa is used as most of the highly preserved ribosomal proteins as well as some housekeeping proteins are of this mass. The method is relatively simple and offers rapid identification (Fig. 1).⁶

Fig. 1.

Mucoid colonies of Mycobacterium abscessus on Lowenstein-Jensen Medium.

Limited work has been done on the use of MALDI-TOF MS for identification of rapidly growing mycobacteria. In this paper we present 35 isolates of rapidly growing mycobacteria from a single outbreak of surgical site infections which were analyzed by using MALDI-TOF MS.

Materials and methods

Thirty five non-repeat isolates of rapidly growing mycobacteria from pus samples of patients suffering from surgical site infections cultured on Lowenstein Jensen Medium were analyzed. Ziehl Neelsen acid fast stain was done on isolates from culture growth to confirm acid fast bacilli (Fig. 2) and routine basic biochemical tests were done to differentiate NTMs from M. tuberculosis complex. The isolates were characterized as rapid growers by growth within 7 days on solid culture media. Further phenotypic identification was done using the following criteria – growth at 43 °C, growth on MacConkey agar, nitrate reduction, growth in 5% NaCl and citrate utilization.⁷

Fig. 2.

Ziehl-Neelsen stain showing acid fast bacilli.

M. abscessus was identified by growth on MacConkey's agar, growth in the presence of 5% NaCl, growth at 37 °C but not at 43 °C and its inability to reduce nitrates or utilize citrate.⁷

The isolates were then analyzed using MALDI-TOF CA (Bruker Daltronics Inc. Billeria MA, USA) (Fig. 3).

Fig. 3.

Bruker MALDI Biotyper System.

The isolates were put through a step of inactivation followed by extraction of proteins. For inactivation, two to three loopful of mycobacteria from Lowenstein-Jensen media were transferred into a 1.5 mL Eppendorf Safe-Lock tube containing 300 μL deionized water. The samples were then heat inactivated by boiling for 30 min.

Protein extraction was carried out using ethanol-acetonitrile. After inactivation, the samples were cooled and 900 μL of absolute alcohol was pipetted into the Eppendorf tube and mixed using a vortex mixer. Centrifugation was carried out at maximum speed (13,000 rpm) and the supernatant decanted. The centrifugation process was repeated and any remaining ethanol was removed. The remaining pellet was allowed to dry at room temperature and a small spatula tip of Zirconia/Silica beads was added. Pure acetonitrile was added to the pellet in amounts ranging from 10 μL to 50 μL according to its size and the tube was vortexed again at maximum speed for 1 min. An equal volume of 70% formic acid was added and vortexed for 5 s followed by centrifugation at 13,000 rpm for 2 min. The supernatant from each of the processed isolates were put on separate spots of the MALDI-TOF steel target plate. Immediately after drying, 1 μL of the MALDI matrix HCCA solution was overlaid. The HCCA contains α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 2.5% trifluoroacetic acid. On a separate spot on the steel target plate 1 μL of a calibration standard was pipetted out and MALDI matrix overlaid.

Spectral analysis

Acquisition of the spectra is done in the range of a mass/charge (m/z) ratio of 2000–20,000. For each isolate the spectra is acquired by summing up of 240 measurements (laser shots). The composite profile which was generated was analyzed using the biotyper software 4.0 using the standard reference library-Mycobacterium v2.0 database. Scoring was done as per criteria established by Bruker Daltonics. Briefly, the spectral patterns obtained are matched and scoring done. Scores of 0–3 are assigned with higher scores showing a higher probability of accurate identification of the organism. Scores greater than 2.0 were considered high (secure species identification), scores of 1.8–2.0 intermediate (secure genus identification), and scores of less than 1.8 were considered unreliable. The software also ranks the ten closest organisms (to species level) in terms of spectra obtained with the isolate (Fig. 4, Fig. 5).

Fig. 4.

Distribution of MALDI–TOF MS scores of Mycobacterium abscessus isolates (n = 35).

Fig. 5.

Mass spectrum of an isolate of Mycobacterium abscessus with a score > 2.0.

Results

All 35 isolates phenotypically and biochemically identified as M. abscessus were further analyzed by MALDI-TOF MS Biotyper. Using the scoring criteria 23 of the 35 isolates (65.7%) were identified with scores of >1.8. In these isolates 9 of the isolates (25.7%) had a secure species identification (Score > 2) while 14 of the isolates had scores between 1.8 and 2.0 which indicated genus-level identification. MALDI biotyper identified M. abscessus as the closest match in the other 12 isolates though the scores obtained were only ranging between 1.418 and 1.613 (Table 1).

Table 1.

MALDI-TOF MS scores of Mycobacterium abscessus isolates.

Most probable organism	MALDI-TOF Score range	No of isolates
Mycobacterium abscessus	1.4–1.6	06
Mycobacterium abscessus	1.6–1.8	06
Mycobacterium abscessus	1.8–2.0	14
Mycobacterium abscessus	2.0–2.2	09
Total		35

The mean score of the isolates identified as M. abscessus was 1.839 with a standard deviation of 0.22522. The 95% confidence interval was 1.76–1.91.

Discussion

M. abscessus is one of the most dreaded nontuberculous mycobacteria and along with Mycobacterium avium is the commonest nontuberculous mycobacteria involved in broncho-pulmonary infection in patients with cystic fibrosis and chronic pulmonary disease.^{8, 9} Outbreaks of skin and soft tissue surgical site infections due to M. abscessus have been reported, demonstrating the threat these organisms pose in causing healthcare-associated infections.^{7, 10}

The American Thoracic Society and the Infectious Diseases Society of America in a joint statement (ATS/IDSA) recommended the identification of NTMs to the species level whenever possible.¹¹ The separation of Mycobacterium chelonae from M. abscessus is clinically important as M. abscessus offers a major threat and is recognized as one of the most resistant organisms to commonly used antibiotics.¹⁰ The resistance to antibiotics could be either intrinsic or acquired by targeted mutation in specific genes in response to the presence of antibiotics.¹⁰

The classification of M. abscessus has been the subject of much debate. The species M. abscessus was sub-classified as three sub species – M. abscessus subsp. abscessus, M. abscessus subsp. massiliense and M. abscessus subsp. bolletii.¹²

However these species have undergone further taxonomical nomenclature changes and presently only two subspecies M. abscessus subsp. bolletii (comprising of the older taxons – M. abscessus subsp. massiliense and M. abscessus subsp. bolletii) and M. abscessus subsp. abscessus are recognized.¹³ Studies suggest that identification to the subspecies level may also be required because of varying resistance patterns.¹⁴

Phenotypic testing for identification of Mycobacteria, has a major disadvantage in that it is unable to separate closely related species.^{5, 15} It also suffers from poor reproducibility and the phenotypic expression may vary within the same species.⁵ A commonly used test for identifying Mycobacteria species is the analysis of the unique mycolic acids present in their cell walls by high-pressure liquid chromatography (HPLC)¹⁶ but this test is not routinely carried out in clinical laboratories. MALDI-TOF MS offers a rapid and reliable method for identification of microorganisms. A wide variety of prokaryotes have been characterized by using this approach. Its use for routine identification of Mycobacteria is, as of yet, not widespread.^{7, 15, 17}

The mycobacterial cell wall is rich in lipids and has peptidoglycans esterified with mycolic acid.¹⁸ Analysis by MALDI-TOF MS requires release of compounds of interest being released from the cell and various methods of extraction have been described^{17, 19} though some authors have described analysis of the whole cell.^{14, 19} The rates of identification to species level in Mycobacteria using the MALDI-TOF MS have been varied.¹⁷ A study reporting the species level identification by using the prescribed scores of Mycobacteria by MALDI-TOF as high as 93.8%(to the species level) has recently been published.²⁰

In our study only 25.7% of isolates had secure species identification by MALDI-TOF MS. However previous studies have indicated that mean confidence scores between 1.7 and 2.0 have yielded similar concordance rates with nucleic acid sequencing as scores greater than 2.0.¹⁶ In our study only 65.7% isolates had scores over 1.7.

Studies have shown an effect of culture time on identification of Mycobacteria by MALDI-TOF with better results shown the earlier the analysis was done.²¹ All our isolates were subcultured 10 days before analysis by MALDI-TOF.

Mycobacterial colony morphology may also have an influence on the quality of spectra identified. Isolates which produce smooth colonies may give poorer quality of spectra resulting in difficulties in identification.¹⁸ Isolates of M. abscessus forming smooth colonies also have a tendency to form clumps and protein extraction is more difficult compared to other nontuberculous mycobacteria.¹⁹ Other factors which may influence the scores obtained are the type of media (liquid versus broth as well as different types of solid media).^{18, 21} Most studies have been done using agar based media or broth while our study was done using growth from Lowenstein Jensen media as this is the commonest media used for culturing Mycobacteria in India.

The advantages of MALDI-TOF MS were that the processing time (including the heat inactivation and extraction steps) was approximately 90 min only, as compared to phenotypic identification, which is growth dependent and takes more than a week to complete and yet is often not reliable. Furthermore, the technique is relatively simple and does not require much expertise. A study has shown a substantial 74% reduction in direct costs after moving to MALDI-TOF MS as compared to traditional molecular techniques.²² The same equipment and matrix can also be used for a wide range of microorganism unlike molecular methods which necessitate the use of specific probes/primers.

The limitations of MALDI-TOF MS are that the databases of spectra of reference organisms are proprietary. In the case of certain organisms refined criteria with lowered cut-off values may be appropriate.²³ The equipment is also presently not widely available.

Several molecular methods including PCR and hybridization assays using specific probes have been developed for identification of Mycobacteria. Commercial kits using hybridization are also available for the identification of nontuberculous mycobacteria. A primary step of DNA amplification is carried out followed by reverse hybridization to oligonucleotide sequences on membranes. These kits are reported to have high sensitivity and specificity.²⁴ However, these methods require technical expertise and take longer (∼6 h).

Conclusion

MALDI-TOF MS offers a rapid, inexpensive method for speciation of nontuberculous mycobacteria. In rapid growers, speciation is of vital importance as treatment guidelines vary among various species and also within a particular species.

MALDI-TOF MS was able to correctly identify all the isolates as M. abscessus though the scores obtained were lesser than previous studies. The spectra quality may be influenced by culture conditions and extraction protocols have to be developed for each laboratory to enhance the spectral quality.

Conflicts of interest

The authors have none to declare.