Three-reaction high-resolution melting assay for rapid differentiation of Mycobacterium tuberculosis complex members.

The possibility of introducing a reliable assay for a quick identification and differentiation of the main species of Mycobacterium tuberculosis complex (MTBC) supports the improvement of efficient tuberculosis combating strategies worldwide. Commercially available assays are often based on cultured samples; however, due to the long cultivation time of mycobacteria, results are delayed. Developed PCR approaches have been published previously, though, when testing intricate veterinary samples, the complex composition of multiplex qPCRs frequently leads to assay failure. In order to overcome those limits, a paradigm of a three-reaction high-resolution melting (HRM) assay for the simultaneous identification and differentiation of the main members of MTBC was established. The assay is based on single nucleotide polymorphisms within gyrB and gyrA, which have been used as target for the establishment of two highly specific HRM assays (HRM assays 1 and 2) discriminating M. tuberculosis/ Mycobacterium canetti, Mycobacterium bovis/M. bovis BCG, Mycobacterium caprae/rare M. caprae/M. bovis ecotypes, Mycobacterium africanum/Mycobacterium orygis/ Mycobacterium pinnipedii/Clade A1, Mycobacterium microti, and a rare subtype of M. canettii followed by a third HRM assay (HRM assay 3) allowing a further differentiation of M. bovis, M. bovis BCG, and a rare subtype of M. caprae/M. bovis, which is considered to be a novel ecotype. High-resolution melting assay 1 is described in a previously published report. High-resolution melting assay 2 showed 100% correlation of all 39 examined isolates with the results of a commercial identification kit. 96% of the clinical samples tested demonstrated concordant results. High-resolution melting assay 3 showed an accordance of 100% with the results of the commercially available identification kit of all 22 samples analyzed. The proposed strategy of the three-reaction HRM assay can be used for an accurate differentiation of up to seven groups of MTBC and potentially to identify a rare subtype of M. canettii either on isolates or on clinical samples.

INTRODUCTION

Tuberculosis is a major cause of human death induced by only one infectious agent resulting in approximately 10 million new infections per year along with about 1.6 million deaths in 2017 (WHO, 2018). Tuberculosis persists as a major health concern not only in humans but also in veterinary medicine.

Mycobacterium tuberculosis complex (MTBC) comprises the closely related species Mycobacterium (M.) tuberculosis, M. bovis, M. bovis Bacillus Calmette and Guérin (BCG), M. caprae, M. africanum, M. microti, M. pinnipedii, M. orygis, and four further species known as animal‐adapted Clade A1 (Dassie bacillus, M. mungi, Chimpanzee bacillus, and M. suricattae) (Brites et al., 2018). Furthermore, Mycobacterium canettii is a genetically more diverse and recombinogenic organism as observed earlier (Fabre et al., 2010), only leading to opportunistic human infections from time to time (Boritsch et al., 2016; Supply et al., 2013). Although its similarity of the nucleotide codes to the species of MBTC, it is not considered to be part of MTBC (Brites et al., 2018). Mycobacterium canettii is mainly limited to the horn of Africa and most of the known strains were isolated in the Republic of Djibouti (Blouin et al., 2014). Mycobacterium tuberculosis is known to be the major source of human tuberculosis; however, numerous cases of infection with other members of the complex are known. Mycobacterium bovis and more rarely M. caprae are the causative agents for bovine tuberculosis, which is recognized to be an important zoonosis responsible for significant economic loss (Rodriguez‐Campos, Smith, Boniotti, & Aranaz, 2014). A recent study (Loiseau et al., 2019) revealed two rare M. caprae/M. bovis ecotypes with no intrinsic pyrazinamide (PZA) resistance, in contrast to the common M. bovis strains, detected in samples from Malawi (Guerra‐Assunção et al., 2015), Germany (Friedrich Loeffler Institute) (defined in this study as ecotype I), and China (Orloski, Robbe‐Austerman, Stuber, Hench, & Schoenbaum, 2018) (defined in this study as ecotype II). These findings resulted in a proposed revision of interpretation of the GenoType MTBC test (Hain Lifescience). In regions where tuberculosis is endemic, neonates were vaccinated with the attenuated M. bovis strain BCG. In immunocompromised children, this procedure can cause a disease pattern similar to the one of tuberculosis (Hesseling et al., 2006). In order to evaluate a zoonotic risk of MTBC, it is important to rely on a fast and accurate method capable of identification and differentiation of the species of MTBC leading to improved programs in public health surveillance and enhanced food safety.

Various molecular assays exist to differentiate species within MTBC. Nevertheless, many approaches have several limits. Most of the methods as, for example, the commercial GenoType MTBC test are not validated for use on clinical samples (Costa, Amaro, et al., 2014; Kasai, Ezaki, & Harayama, 2000; Niemann, Harmsen, Rüsch‐Gerdes, & Richter, 2000; Pinsky & Banaei, 2008; Pounder et al., 2010; Reddington et al., 2012). Since Mycobacteria require many months to obtain considerable growth in culture, methods based on cultivated isolates require much time to determine the correct MTBC species. Other methods are very laborious (Kamerbeek et al., 1997), require expensive equipment (Jagielski et al., 2014), or rely on multiplex qPCR assays (Costa, Amaro, et al., 2014; Halse, Escuyer, & Musser, 2011; Pinsky & Banaei, 2008; Pounder et al., 2010; Reddington et al., 2012). However, the complex composition of a multiplex qPCR can often lead to PCR inhibition due to intricate organ samples of animal tissues.

High‐resolution melting (HRM) approaches are cheap and rapid assays, which are able to detect single nucleotide polymorphism (SNP) according to altered melting temperatures (T) of dissociating PCR amplicons (Vossen, Aten, Roos, & Dunnen, 2009). A fluorescent nucleic acid dye is intercalating with the resulting PCR amplicons, which are dissociating upon increase in temperature and thus resulting in a decrease in fluorescence intensity. The determinate of T is based on its nucleotide sequence, length, and level of GC. The user‐friendly single‐plex HRM assay can be completed within roughly 2 hr. Moreover, as a major advantage, HRM assays can be performed using samples directly extracted from clinical tissue. High‐resolution melting assays have been used to discriminate various bacteria species (Esteves et al., 2018; Jeffery, Gasser, Steer, & Noormohammadi, 2007; Robertson et al., 2009; Stephens, Inman‐Bamber, Giffard, & Huygens, 2008; Winchell, Wolff, Tiller, Bowen, & Hoffmaster, 2010) or to analyze antibiotic resistance in M. tuberculosis (Anthwal et al., 2017; Chen et al., 2011; Yadav et al., 2012). Moreover, HRM assays have been established for differentiation of nontuberculous mycobacteria (NTM) and confining them from MTBC (Issa et al., 2014; Khosravi, Hashemzadeh, Hashemi Shahraki, & Teimoori, 2017; Perng et al., 2012). Some studies combined HRM with multiplex qPCR assays targeting at the region of difference (RD) (Pinsky & Banaei, 2008; Pounder et al., 2010).

We have previously reported the design and evaluation of a HRM assay (HRM assay 1) for the identification and differentiation of MTBC into three groups most relevant for veterinarians (Landolt, Stephan, & Scherrer, 2019). In the present study, two additional HRM assays (HRM assays 2 and 3) were developed with the aim to discern the main members of MTBC. By combining HRM assays 1 and 2 targeting six SNPs on the gyrB gene, a two‐step paradigm was obtained, distinguishing M. tuberculosis/M. canettii, M. canettii (rare subtype), M. africanum/M. orygis/M. pinnipedii/Clade A1, M. bovis/M. bovis BCG, M. caprae/rare M. caprae/M. bovis ecotypes, and M. microti in cultured isolates and clinical samples. If requested, it is possible to further differentiate M. bovis, M. bovis BCG, and one of the two rare M. caprae/M. bovis ecotypes (ecotype I) by conducting another HRM assay (HRM assay 3) based on two SNPs within gyrA and thus completing the three‐step paradigm. For most diagnostic applications, however, a combination of HRM assays 1 and 2 will be sufficient.

MATERIALS AND METHODS

Samples and reference strains

Sixty‐one samples positive for MTBC were collected from 39 different animals (Table 1, Table A1). One wild boar isolate was put at our disposal by courtesy of Lucía de Juan Ferré and Beatriz Romero Martínez. In total, 62 samples including 39 cultured isolates and 23 directly extracted clinical samples were tested with HRM assay 2. High‐resolution melting assay 3 was validated to distinguish between M. bovis and M. bovis BCG using 15 M. bovis isolates and 7 M. bovis clinical samples. Positive control samples (M. microti ATCC 19422, M. bovis BCG Pasteur ATCC 35734, M. bovis BCG Tice ATCC 27289, and M. tuberculosis H37Rv) were used in each qPCR run. Clinical samples were received in the Laboratory of Veterinary Bacteriology, University of Zurich, between 2013 and 2016.

Table 1

MTBC‐positive samples used for the development of the HRM assays 2 and 3

Species	Host	HRM assay 2	HRM assay 3
		No. of isolates	No. of isolates
Cultured material (n = 39)
M. tuberculosis	Elephant	3
M. caprae	Cow	7
M. bovis	Cow	15	15
M. microti	Cat	8
M. microti	Alpaca	3
M. microti	Llama	2
M. microti	Wild boar (Spain)	1
Clinical samples (n = 23)
M. tuberculosis	Elephant	2
M. caprae	Cow	4
M. bovis	Cow	7	7
M. microti	Cat	5
M. microti	Alpaca	3
M. microti	Llama	2
Total		62	22

Thirty‐nine isolates obtained from cultured material, whereas 23 samples were clinical samples directly extracted from tissue samples. Sixty‐one samples were derived from Switzerland, whereas one isolate originated from Spain.

Table A1

MTBC‐positive isolates used for the development of the high‐resolution melting (HRM) assays 2 and 3

Sample	Species	Origin	Specimen	Culture?	Clinical specimen?
M. bovis (15)
20175	M. bovis	Cow	Lung	Cultured	Not available
20482	M. bovis	Cow	Lung, lymph node pool	Cultured	Not available
20531	M. bovis	Cow	Lymph node pool	Cultured	Not available
20593	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20594	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20596	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20597	M. bovis	Cow	Lymph node pool	Cultured	Not available
20599	M. bovis	Cow	Lymph node pool	Cultured	Not available
20600	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20606	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20608	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20609	M. bovis	Cow	Lymph node pool	Cultured	Clinical specimen
20665	M. bovis	Cow	Lymph node pool	Cultured	Not available
22539	M. bovis	Cow	Lymph node	Cultured	Not available
22667	M. bovis	Cow	Lymph node	Cultured	Not available
M. caprae (7)
14‐13	M. caprae	Cow	Lymph node	Cultured	Not available
22914	M. caprae	Cow	Lymph node pool	Cultured	Clinical specimen
22971	M. caprae	Cow	Lymph node pool	Cultured	Clinical specimen
22848	M. caprae	Cow	Liver, lung, lymph node pool	Cultured	Clinical specimen
22966	M. caprae	Cow	Lymph node pool	Cultured	Not available
13‐450	M. caprae	Cow	Lymph node	Cultured	Not available
13‐162	M. caprae	Cow	Lymph node	Cultured	Clinical specimen
M. microti (15)
22928	M. microti	Cat	Lung, lymph node pool	Cultured	Clinical specimen
15‐1765	M. microti	Cat	Lung	Cultured	Clinical specimen
15‐342	M. microti	Alpaca	Lymph node	Cultured	Clinical specimen
14‐58	M. microti	Cat	Lung	Cultured	Clinical specimen
14‐690	M. microti	Alpaca	Liver	Cultured	Clinical specimen
17‐2287	M. microti	Alpaca	Spleen	Cultured	Not available
15817	M. microti	Lama	Lymph node	Cultured	Clinical specimen
15‐1955	M. microti	Cat	Lymph node	Cultured	Clinical specimen
16‐2156	M. microti	Cat	Bronchoalveolar lavage	Cultured	Not available
1522744	M. microti	Cat	Lung, lymph node pool	Cultured	Not available
16‐1347	M. microti	Cat	Lung	Cultured	Not available
17‐1084	M. microti	Cat	Lymph node, skin	Cultured	Clinical specimen
17‐1063	M. microti	Alpaca	Lymph node, lung, heart, liver, cervical vertebra	Culture ongoing	Clinical specimen
17‐549	M. microti	Lama	Liver	Cultured	Clinical specimen
MI16	M. microti	Wildboar Spain	Unknown	Cultured	Not available
M. tuberculosis (4)
15‐961‐2	M. tuberculosis	Elephant 1	Lung	Cultured	Clinical specimen
15‐961‐1	M. tuberculosis	Elephant 1	Pharyngeal swab	Not cultivated	Clinical specimen
15‐1115‐2	M. tuberculosis	Elephant 2	Lung	Cultured	Not available
15‐1221‐1	M. tuberculosis	Elephant 3	Lung	Cultured	Not available

Culture and DNA extraction

Sample preparation, culture, and DNA extraction were conducted as described in a former study (Ghielmetti et al., 2017). GenoType MTBC test (Hain Lifescience), spoligotyping (Ruettger et al., 2012), and multilocus variable number tandem repeat analysis using an internationally established 24‐loci panel (Supply et al., 2006) were used for species identification of cultured isolates. Standard biosecurity procedures were followed for handling of samples.

HRM development

High‐resolution melting development of assay 1 was described in a former study (Landolt et al., 2019). The differentiation of the main members of MTBC is performed applying a paradigm of a three‐reaction HRM assay based on four SNPs located on gyrB (base pair positions 675, 756, 1,410, and 1,450 [Niemann et al., 2000] and one SNP on gyrA [base pair position 1,323]). In order to extend the obtained SNP differentiation scheme with the recently described two rare ecotypes of M. caprae/M. bovis (ecotypes I and II), a rare M. canettii subtype (Loiseau et al., 2019) and the animal‐adapted MTBC clades (Brites et al., 2018), two additional SNPs on gyrB (base pair positions 1,437 and 1,439), and one SNP on gyrA (base pair position 1,359) were included in the assay design (Figure 1).

Figure 1

Differentiation of MTBC based on eight different single nucleotide polymorphisms (SNPs). Numbers represent the position of the SNP in relation to the start codon of gyrB and gyrA, respectively. (a) Six SNPs on gyrB are represented. HRM assay 1 (Landolt et al., 2019) allows the distinction between M. tuberculosis/M. canettii/M. africanum/M. orygis/M. pinnipedii/Clade A1, M. microti/M. canettii (rare subtype) and M. bovis/M. bovis BCG/M. caprae/rare M. caprae/M. bovis ecotypes. HRM assay 2 can differentiate between M. tuberculosis/M. canettii, M. canettii (rare subtype), M. africanum/M. orygis/M. pinnipedii/Clade A1/M. microti/M. caprae/rare M. caprae/M. bovis ecotypes, and M. bovis/M. bovis BCG. Combining the results of HRM assays 1 and 2, six groups can be distinguished: M. canettii (rare subtype), M. tuberculosis/M. canettii, M. africanum/M. orygis/M. pinnipedii/Clade A1, M. microti, M. caprae/rare M. caprae/M. bovis ecotypes, and M. bovis/M. bovis BCG. (b) Two SNPs on positions 1,323 and 1,359 of gyrA are illustrated indicating the differentiation between M. bovis, M. bovis BCG, and rare M. caprae/M. bovis ecotype I. *Rare subtype, highly recombinogenic strain, intrinsic pyrazinamide (PZA) resistance. **Intrinsic PZA resistance. ***No intrinsic PZA resistance. ****Intrinsic PZA and cycloserine resistance

Primer of HRM assays

Primers were designed based on alignments of the available sequences of gyrB and gyrA from MTBC (GenBank accession numbers: gyrB M. bovis AB014184.1, gyrB M. microti AB014205.1, gyrB M. tuberculosis X78888.1, gyrB M. africanum FR878060.1, gyrB M. caprae CP016401.1, gyrA M. bovis LT708304.1, and gyrA M. bovis BCG Pasteur AM408590.1). The designed primers amplify a conserved 107‐base pair (bp) fragment within gyrB comprising two SNPs and gyrA comprising one SNP, respectively (Figure 1). Gene specificity of all primers was confirmed by NCBI BLAST searches. Primers were HPLC‐purified and synthesized by Microsynth. Three primer pairs were designed for the three‐reaction HRM paradigm (Table 2).

Table 2

Primers used for the three‐reaction HRM paradigm

Assay	Primer	Target	Amplicon	Primer sequence (5′ → 3′)	Reference
HRM assay 1	HRM_gyrB_for HRM_gyrB_rev	gyrB	144 bp	CGGCTCGAAGTCGAGATCAAG TTCGAAAACAGCGGGGTCG	Landolt et al. (2019)
HRM assay 2	HRM_gyrB2_for HRM_gyrB2_rev	gyrB	107 bp	CAAATCGTTTGTGCAGAAGGTCTG CTTGCGCCGAGGACACAG	This study
HRM assay 3	HRM_gyrA_for HRM_gyrA_rev	gyrA	107 bp	AGGCAATCCTGGACATGCAG GATGTCTTCCAGATCGGCGATC	This study

qPCR and melting conditions

Each HRM assay was processed separately on a Rotor‐Gene Q system (Qiagen) with the Type‐it HRM PCR Kit (Qiagen). The qPCRs were performed as described previously (Landolt et al., 2019). High‐resolution melting ramping from 76°C to 93°C was applied. Fluorescence data were measured every 2 s at 0.1°C increments generating specific melting curves. Reference strains M. microtii ATCC 19422, M. bovis BCG Pasteur ATCC 35734, and M. tuberculosis H37Rv were used as melting curve standards and positive controls. Additionally, for HRM assay 3, M. bovis BCG Tice ATCC 27289 was used. As a negative control, ultrapure water was included in each experiment. Rotor‐Gene Q Software 2.3.1 (Qiagen) was used for data analysis to generate normalized and difference plots as described in a previous study (Landolt et al., 2019). To prevent false‐negative results possibly deriving from inhibition, clinical samples were analyzed in duplicate undiluted and in the form of a 1:5 dilution. The cultured isolates were analyzed at concentrations between 100 pg and 10 ng.

To investigate the intra‐ and interassay variability of the T, illustrating the repeatability of the developed HRM assays, a randomly chosen subset of 22 cultured isolates and 18 clinical specimens for HRM assays 2 and 9 cultured isolates and 7 clinical specimens for HRM assay 3 were tested, respectively. The variability assays were conducted in triplicates in three single runs at three different days.

Specificity

To check for possible nonspecific signals of HRM assays 2 and 3, 41 different NTM, Nocardia paucivorans, Escherichia coli, and Streptococcus suis were tested (Table A2).

Table A2

Exclusivity panel consisting of 41 nontuberculous mycobacteria and 3 nonmycobacterial species for specificity testing of the high‐resolution melting (HRM) assays 2 and 3

Species	No. of isolates	Species	No. of isolates
M. abscessus sp.	2	M. malmoense	1
M. avium subsp. avium	2	M. marinum	1
M. avium subsp. hominissuis	32	M. monacense	3
M. avium subsp. paratuberculosis	1	M. nebraskense	1
M. avium subsp. silvaticum	1	M. neoaurum	5
M. bourgelatii	1	M. nonchromogenicum	7
M. celatum	1	M. palustre	1
M. chelonae subsp. chelonae	1	M. parafortuitum	2
M. chimaera/intracellulare/youngonense	1	M. paragordonae	6
M. chitae	1	M. peregrinum	2
M. elephantis	1	M. persicum	2
M. engbaekii	1	M. phlei	3
M. europaeum	1	M. scrofulaceum	1
M. fortuitum/porcinum	1	M. simiae	1
M. goodii	1	M. smegmatis	1
M. gordonae	2	M. szulgai	1
M. hassiacum	1	M. terrae	1
M. interjectum/paraense	1	M. vaccae	4
M. intermedium	1	M. xenopi	5
M. intracellulare	3	Nocardia paucivorans	1
M. kansasii	10	Escherichia coli	1
M. lymphaticum	1	Streptococcus suis	1

Sensitivity

The analytical sensitivity of HRM assay 2 was measured by triplicate testing of a 10‐fold dilution series of DNA isolated from strains M. microti ATCC 19422, M. bovis BCG Pasteur ATCC 35734, M. caprae clinical isolate ZH22914, and M. tuberculosis H37Rv with known concentrations in genome equivalents (GE). The analytical sensitivity of HRM assay 3 was tested for M. bovis BCG Pasteur ATCC 35734 and the clinical isolate M. bovis ZH20665 in an analogous manner by analyzing 10‐fold DNA dilution series. Based on an estimated genome size of 4.4 Mb for members of MTBC, 1 GE correlates with a DNA quantity of 4.8 fg. The slope of the resulting standard curve corresponded to the amplification efficiency of each tested strain. The limit of detection (LOD) was defined as lowest dilution for samples with successful PCR amplification of all triplicates having a Ct < 38 and a standard deviation of ≤ 0.5.

RESULTS

HRM of cultured isolates

All cultured isolates tested were amplified successfully yielding a melting curve. The resulting species‐specific T (Tables 3 and 4, Tables A3 and A4) of corresponding melting curves (Figures 2 and 3) from sample subsets used for the determination of the intra‐ and interassay variability clearly represent three independent groups in the case of HRM assay 2 and two distinct groups with HRM assay 3, respectively. Obtained T ranges between the different groups are very close to each other impeding a clear differentiation between members of MTBC solely based on T. In contrast, the illustration of dissociating curves in the form of difference plots (Figures 2c and 3c) as well as normalized plots (Figures 2b and 3b) allowed a clear differentiation between MTBC species.

Table 3

Intra‐ and interassay variability of HRM assay 2 of cultured samples

	Run 1		Run 2		Run 3		Interassay
	Tm	CV%	Tm	CV%	Tm	CV%	Tm	CV%
M. tuberculosis H37Rv	83.55		83.75		83.6		83.63	0.12
M. bovis BCG Pasteur ATCC 35734	82.62		82.77		82.70		82.7	0.09
M. microti ATCC 19422	83.13		83.25		83.28		83.22	0.10
M. tuberculosis (n = 3)	83.56 ± 0.04	0.01	83.84 ± 0.04	0.02	83.66 ± 0.06	0.04	83.70 ± 0.18	0.17
M. bovis (n = 6)	82.70 ± 0.10	0.04	82.93 ± 0.08	0.03	82.74 ± 0.06	0.02	82.80 ± 0.20	0.14
M. caprae (n = 6)	83.12 ± 0.09	0.03	83.27 ± 0.09	0.02	83.23 ± 0.06	0.02	83.19 ± 0.16	0.11
M. microti (n = 7)	83.10 ± 0.08	0.02	83.34 ± 0.09	0.04	83.29 ± 0.04	0.01	83.23 ± 0.21	0.14

Mean values and standard deviation of melting temperatures (T) of a randomly chosen subset of cultured samples are listed. Corresponding coefficients of variation (CV) in % are indicated for each MTBC species tested.

Table 4

Intra‐ and interassay variability of HRM assay 3 of cultured samples

	Run 1		Run 2		Run 3		Interassay
	Tm	CV%	Tm	CV%	Tm	CV%	Tm	CV%
M. tuberculosis H37Rv	86.82		86.82		86.85		86.83	0.02
M. microti ATCC 19422	86.83		86.83		86.83		86.83	0
M. bovis BCG Pasteur ATCC 35734	86.42		86.38		86.42		86.41	0.03
M. bovis (n = 7)	86.83 ± 0.50	0.01	86.83 ± 0.08	0.02	86.83 ± 0.10	0.03	86.83 ± 0.10	0.03
M. bovis BCG (n = 2)	86.39 ± 0.06	0.03	86.41 ± 0.03	0.03	86.45 ± 0.03	0.03	86.41 ± 0.08	0.04

Mean values and standard deviation of melting temperatures (T) of a randomly chosen subset of cultured samples are listed. Corresponding coefficients of variation (CV) in % are indicated for each MTBC species tested.

Table A3

Raw data set and statistical parameters generated from the intra‐ and interassay variability of high‐resolution melting (HRM) assay 2 using a randomly chosen subset of 22 cultured samples

Isolate			Run 1						Run 2						Run 3						Inter‐Assay
MTBC member	Sample no	Origin	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Mean Tm	SD	CV%	Inter‐assay CV%
M. microti	ATCC 19422		20.65	83.13					20.87	83.25					19.74	83.28					83.22	0.08	0.10
M. bovis BCG Pasteur	ATCC 35734		24.67	82.62					24.82	82.77					23.57	82.70					82.70	0.08	0.09
M. tuberculosis	H37RV		22.41	83.55					22.44	83.75					20.97	83.60					83.63	0.10	0.12
M. tuberculosis	15‐961 2B	Elephant	14.87	83.60	83.58	0.02	0.02		15.11	83.80	83.80	0.00	0.00		14.01	83.70	83.69	0.02	0.02		83.69	0.11	0.13
			14.95	83.58					15.13	83.80					14.05	83.70
			14.96	83.57					15.14	83.80					13.94	83.67
M. tuberculosis	15‐1115‐2	Elephant	16.34	83.55	83.54	0.01	0.01		16.69	83.87	83.86	0.03	0.03		15.32	83.63	83.65	0.03	0.03		83.69	0.16	0.19
			16.71	83.55					16.65	83.83					15.47	83.65
			16.73	83.53					16.75	83.88					15.46	83.68
M. tuberculosis	15‐1221‐1	Elephant	19.59	83.53	83.53	0.01	0.01	0.01	19.98	83.87	83.83	0.04	0.04	0.02	18.45	83.60	83.67	0.06	0.07	0.04	83.68	0.15	0.18	0.17
			19.80	83.52					20.02	83.83					18.45	83.68
			19.77	83.53					19.75	83.80					18.47	83.72
M. caprae	229171	Cow	20.74	83.05	83.05	0.00	0.00		21.20	83.23	83.24	0.01	0.01		19.79	83.23	83.23	0.02	0.02		83.17	0.11	0.13
			37.30						21.09	83.25					19.76	83.22
			21.07	83.05					21.28	83.23					19.55	83.25
M. caprae	22914	Cow	20.60	83.10	83.12	0.03	0.03		20.97	83.30	83.26	0.04	0.04		19.52	83.27	83.25	0.02	0.02		83.21	0.08	0.10
			20.68	83.10					21.02	83.25					19.59	83.25
			20.71	83.15					20.79	83.23					19.48	83.23
M. caprae	22848	Cow	22.93	83.07	83.04	0.02	0.03		23.24	83.20	83.19	0.01	0.01		21.68	83.23	83.20	0.03	0.04		83.14	0.09	0.10
			22.94	83.03					23.20	83.18					21.77	83.20
			22.90	83.03					23.34	83.18					21.53	83.17
M. caprae	22966	Cow	19.89	83.13	83.15	0.02	0.02		20.04	83.35	83.33	0.04	0.05		18.60	83.25	83.26	0.02	0.02		83.25	0.09	0.11
			19.84	83.15					19.97	83.35					18.46	83.25
			19.81	83.17					20.04	83.28					18.84	83.28
M. caprae	13‐450	Cow	18.04	83.20	83.16	0.04	0.04		18.23	83.30	83.31	0.02	0.02		17.00	83.28	83.28	0.01	0.01		83.25	0.08	0.09
			18.04	83.13					18.10	83.33					16.99	83.28
			18.13	83.15					18.21	83.30					16.93	83.27
M. caprae	13‐162	Cow	22.08	83.13	83.10	0.03	0.03	0.03	22.43	83.23	83.24	0.01	0.01	0.03	20.91	83.25	83.26	0.01	0.01	0.02	83.20	0.08	0.10	0.11
			21.93	83.10					22.23	83.25					20.75	83.25
			22.01	83.08					22.33	83.23					20.87	83.27
M. bovis	20593	Cow	26.50	82.65	82.64	0.01	0.01		27.13	82.88	82.93	0.06	0.08		25.69	82.68	82.69	0.01	0.01		82.75	0.15	0.18
			24.89	82.65					23.85	83.00					24.93	82.70
			25.63	82.63					24.38	82.90					25.46	82.70
M. bovis	20594	Cow	25.56	82.60	82.61	0.02	0.02		26.03	82.88	82.88	0.03	0.03		24.80	82.72	82.71	0.02	0.03		82.73	0.14	0.16
			24.81	82.60					25.75	82.85					24.77	82.72
			25.43	82.63					25.79	82.90					23.95	82.68
M. bovis	20175	Cow	18.77	82.80	82.74	0.05	0.06		18.90	82.98	82.93	0.07	0.09		17.68	82.80	82.78	0.03	0.03		82.82	0.10	0.12
			18.80	82.70					18.92	82.97					17.53	82.80
			18.81	82.73					18.94	82.85					17.66	82.75
M. bovis	20596	Cow	21.91	82.67	82.67	0.08	0.09		22.37	82.95	82.94	0.01	0.01		21.14	82.80	82.76	0.04	0.04		82.79	0.14	0.17
			22.42	82.75					22.47	82.93					20.81	82.75
			22.56	82.60					22.43	82.95					21.01	82.73
M. bovis	22667	Cow	19.91	82.80	82.80	0.00	0.00		20.04	82.93	82.92	0.01	0.01		18.95	82.78	82.78	0.01	0.01		82.83	0.08	0.10
			19.84	82.80					20.05	82.92					18.79	82.77
			19.81	82.80					20.05	82.92					19.09	82.78
M. bovis	22539	Cow	21.42	82.73	82.76	0.04	0.04	0.04	21.80	82.92	82.92	0.00	0.00	0.04	20.49	82.77	82.76	0.01	0.01	0.02	82.81	0.09	0.11	0.14
			21.55	82.75					21.44	82.92					20.44	82.75
			21.38	82.80					21.60	82.92					20.52	82.75
M. microti	22928	Cat	16.90	83.17	83.17	0.02	0.02		17.13	83.40	83.42	0.02	0.02		16.01	83.30	83.31	0.01	0.01		83.30	0.13	0.15
			16.93	83.18					17.13	83.43					15.89	83.30
			17.01	83.15					17.08	83.43					16.13	83.32
M. microti	15‐1765	Cat	21.77	83.18	83.16	0.03	0.03		22.21	83.32	83.30	0.02	0.02		20.79	83.28	83.28	0.02	0.02		83.25	0.08	0.09
			21.44	83.17					21.93	83.30					19.99	83.27
			21.78	83.13					22.05	83.28					20.59	83.30
M. microti	14‐58	Cat	24.40	83.02	83.02	0.01	0.01		24.39	83.28	83.27	0.02	0.02		23.44	83.28	83.28	0.02	0.02		83.19	0.15	0.18
			24.39	83.02					24.64	83.28					23.18	83.30
			24.31	83.03					24.43	83.25					23.45	83.27
M. microti	14‐690	Alpaca	15.10	83.05	83.05	0.03	0.03		15.33	83.30	83.29	0.01	0.01		14.28	83.30	83.29	0.01	0.01		83.21	0.14	0.17
			15.13	83.07					15.27	83.28					14.20	83.30
			15.21	83.02					15.29	83.30					14.24	83.28
M. microti	16‐2156	Cat	23.07	83.07	83.08	0.02	0.02		23.48	83.30	83.31	0.01	0.01		22.32	83.27	83.27	0.00	0.00		83.22	0.12	0.14
			23.49	83.08					23.60	83.30					22.17	83.27
			23.49	83.10					23.57	83.32					22.01	83.27
M. microti	1522744	Cat	22.79	83.12	83.09	0.03	0.03		22.89	83.33	83.29	0.03	0.04		21.82	83.27	83.26	0.01	0.01		83.21	0.11	0.13
			22.67	83.07					22.97	83.27					21.74	83.27
			22.68	83.08					22.94	83.27					21.49	83.25
M. microti	16‐1347	Cat	18.92	83.13	83.15	0.03	0.03	0.02	19.20	83.30	83.27	0.03	0.03	0.02	17.52	83.30	83.30	0.00	0.00	0.01	83.24	0.08	0.10	0.14
			18.94	83.13					18.99	83.25					17.21	83.30
			18.87	83.18					18.86	83.25					17.32	83.30

Table A4

Raw data set and statistical parameters generated from the intra‐ and interassay variability of high‐resolution melting (HRM) assay 3 using a randomly chosen subset of 9 cultured samples and 7 clinical specimens

Isolate			Run 1						Run 2						Run 3						Inter‐Assay
MTBC member	Sample no	Origin	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Mean Tm	SD	CV%	Inter‐assay CV%
M. microti	ATCC 19422		26.37	86.83					28.03	86.83					28.56	86.83					86.83	0.00	0.00
M. tuberculosis	H37RV		29.35	86.82					29.36	86.82					29.03	86.85					86.83	0.02	0.02
M. bovis isolate	20593	Cow	35.04	86.88	86.86	0.02	0.02		33.95	86.85	86.85	0.00	0.00		33.73	86.87	86.90	0.03	0.03		86.87	0.02	0.03
			35.10	86.85					33.59	86.85					33.52	86.92
			35.36	86.85					33.70	86.85					32.61	86.90
M. bovis isolate	20594	Cow	32.92	86.85	86.85	0.00	0.00		32.73	86.85	86.86	0.02	0.02		32.36	86.83	86.88	0.05	0.05		86.86	0.02	0.02
			33.48	86.85					32.56	86.85					32.62	86.90
			33.28	86.85					32.83	86.88					32.60	86.92
M. bovis isolate	20609	Cow	35.18	86.83	86.83	0.02	0.02		34.71	86.87	86.87	0.03	0.03		33.73	86.90	86.87	0.06	0.07		86.86	0.02	0.02
			34.74	86.85					33.89	86.85					33.87	86.80
			34.54	86.82					34.39	86.90					33.60	86.90
M. bovis isolate	20608	Cow	28.92	86.82	86.82	0.01	0.01		28.69	86.83	86.84	0.01	0.01		27.96	86.78	86.77	0.04	0.04		86.81	0.04	0.04
			28.55	86.83					28.50	86.85					28.36	86.73
			29.49	86.82					28.75	86.85					28.46	86.80
M. bovis isolate	20596	Cow	29.93	86.78	86.79	0.01	0.01		29.35	86.83	86.84	0.01	0.01		29.37	86.80	86.80	0.00	0.00		86.81	0.02	0.03
			29.69	86.80					29.25	86.85					28.64	86.80
			29.87	86.80					29.30	86.83					28.92	86.80
M. bovis isolate	20600	Cow	27.83	86.83	86.83	0.00	0.00		29.51	86.85	86.83	0.02	0.02		29.82	86.85	86.84	0.01	0.01		86.84	0.01	0.01
			29.66	86.83					30.07	86.83					29.91	86.83
			30.09	86.83					30.49	86.82					29.91	86.85
M. bovis isolate	20606	Cow	28.36	86.80	86.81	0.02	0.02	0.01	27.94	86.80	86.78	0.03	0.03	0.02	27.42	86.83	86.84	0.01	0.01	0.03	86.81	0.03	0.03	0.03
			27.86	86.80					28.01	86.78					27.34	86.85
			28.12	86.83					27.88	86.75					27.47	86.83
M. bovis specimen	20593	Cow	19.04	86.83	86.84	0.01	0.01		18.82	86.82	86.80	0.02	0.02		18.64	86.88	86.92	0.04	0.04		86.85	0.06	0.07
			18.84	86.85					19.19	86.80					18.80	86.92
			18.65	86.85					18.95	86.78					18.68	86.95
M. bovis specimen	20594	Cow	36.86		86.88	0.01	0.01		16.22	86.78	86.78	0.02	0.02		16.16	86.90	86.91	0.01	0.01		86.86	0.06	0.07
			16.18	86.88					16.51	86.80					16.18	86.92
			16.75	86.87					16.55	86.77					16.07	86.90
M. bovis specimen	20609	Cow	29.46	86.85	86.85	0.00	0.00		28.99	86.80	86.82	0.03	0.03		28.66	86.88	86.85	0.03	0.03		86.84	0.02	0.02
			29.75	86.85					29.35	86.80					28.51	86.85
			29.04	86.85					29.45	86.85					28.57	86.83
M. bovis specimen	20608	Cow	17.18	86.85	86.88	0.03	0.03		17.34	86.85	86.86	0.01	0.01		17.03	86.90	86.88	0.03	0.03		86.87	0.01	0.02
			17.43	86.90					17.60	86.85					17.13	86.90
			17.14	86.88					17.64	86.87					17.22	86.85
M. bovis specimen	20596	Cow	16.60	86.85	86.86	0.04	0.05		16.99	86.90	86.90	0.00	0.00		16.66	86.87	86.86	0.01	0.01		86.87	0.03	0.03
			17.02	86.90					17.01	86.90					16.54	86.85
			16.84	86.82					17.14	86.90					16.42	86.85
M. bovis specimen	20600	Cow	16.88	86.85	86.85	0.00	0.00		17.15	86.88	86.88	0.03	0.03		16.97	86.85	86.84	0.01	0.01		86.86	0.02	0.02
			16.83	86.85					17.47	86.90					16.96	86.85
			16.74	86.85					17.24	86.85					16.64	86.83
M. bovis specimen	20606	Cow	17.22	86.85	86.85	0.00	0.00	0.01	17.18	86.85	86.85	0.02	0.02	0.02	16.72	86.85	86.83	0.03	0.03	0.02	86.84	0.01	0.02	0.04
			17.34	86.85					17.31	86.83					16.77	86.80
			16.90	86.85					17.23	86.87					16.87	86.83
M. bovis BCG Tice isolate	ATCC 27289		30.43	86.40	86.37	0.04	0.04		30.70	86.40	86.40	0.02	0.02		30.26	86.43	86.44	0.01	0.01		86.40	0.04	0.04
			30.84	86.38					30.51	86.38					30.14	86.45
			30.73	86.33					30.22	86.42					29.81	86.45
M. bovis BCG Pasteur isolate	ATCC 35734		30.62	86.45	86.44	0.02	0.02	0.03	31.32	86.42	86.41	0.03	0.03	0.03	30.56	86.47	86.46	0.03	0.04	0.03	86.44	0.02	0.03	0.04
			30.72	86.42					31.15	86.38					30.23	86.42
			30.78	86.45					31.14	86.43					30.27	86.48

Figure 2

Representative high‐resolution melting graphs corresponding to one high‐resolution melting analysis of a subset of cultured samples (n = 22) of HRM assay 2. Curves of tested samples previously identified as M. tuberculosis are shown in yellow, M. microti in blue, M. bovis/M. bovis BCG in red, and M. caprae in green. (a) Melting curves; (b) normalized plot; and (c) difference plot in relation to M. bovis BCG Pasteur ATCC 35734

Figure 3

Representative high‐resolution melting graphs corresponding to one high‐resolution melting analysis of a subset of cultured samples (n = 9) and clinical specimens (n = 7) of HRM assay 3. Curves of tested samples previously identified as M. bovis BCG are shown in pink, cultured samples of M. bovis in red, and clinical specimen of M. bovis in orange. (a) Melting curves; (b) normalized plot; and (c) difference plot in relation to M. bovis BCG Pasteur ATCC 35734

High‐resolution melting assay 2 differentiates three groups, namely M. tuberculosis, M. microti/M. caprae, and M. bovis/M. bovis BCG. The obtained intra‐assay coefficients of variation (CVs) and the interassay CVs were ranging between 0.01%–0.04% and 0.11%–0.17%, respectively (Table 3, Table A3), demonstrating highly reproducible and robust assays. Species identification results of all 39 (100%) tested cultured isolates were in accordance with the species classification of GenoType MTBC test (Hain Lifescience) results.

High‐resolution melting assay 3 differentiates M. bovis from M. bovis BCG. The intra‐assay CVs and the interassay CVs yielded T values ranging between 0.01%–0.03% and 0.03%–0.04%, respectively (Table 4, Table A4). Species identification results of all 15 (100%) tested cultured isolates were in agreement with the GenoType MTBC test (Hain Lifescience) results.

HRM of clinical samples

For 22/23 (96%) clinical samples tested in HRM assay 2, obtained normalized and difference plots (Figure 4b,c) showed the appearance of three distinct groups, namely M. tuberculosis, M. microti/M. caprae, and M. bovis/M. bovis BCG in accordance with the results of GenoType MTBC test (Hain Lifescience). One sample revealed significantly lower T values compared with the other results. This clinical sample (samples 17–1,063), however, has not yet been successfully cultured and is in the progress of further investigations. High‐resolution melting assay 3 showed a clear distinction of all cultured isolates as well as directly isolated clinical samples of M. bovis from M. bovis BCG in 100% concordance with the GenoType MTBC test (Hain Lifescience) results.

Figure 4

Representative high‐resolution melting graphs corresponding to one high‐resolution melting analysis of a subset of clinical specimens (n = 18) for HRM assay 2. Curves of tested samples previously identified as M. tuberculosis are shown in yellow, M. microti in blue, M. bovis/M. bovis BCG in red, and M. caprae in green. (a) Melting curves; (b) normalized plot; and (c) difference plot in relation to M. bovis BCG Pasteur ATCC 35734

In HRM assay 2, intra‐ and interassay CVs were between 0.02%–0.04% and between 0.12%–0.19%, respectively (Table 5, Table A5).

Table 5

Intra‐ and interassay variability of HRM assay 2 of clinical specimens

	Run 1		Run 2		Run 3		Interassay
	Tm	CV%	Tm	CV%	Tm	CV%	Tm	CV%
M. tuberculosis H37Rv	83.85		83.67		83.58		83.70	0.16
M. bovis BCG Pasteur ATCC 35734	82.82		82.73		82.62		82.72	0.12
M. microti ATCC 19422	83.30		83.25		83.08		83.21	0.14
M. tuberculosis (n = 2)	83.71 ± 0.13	0.03	83.60 ± 0.10	0.03	83.52 ± 0.09	0.03	83.63 ± 0.20	0.12
M. bovis (n = 6)	82.97 ± 0.10	0.04	82.82 ± 0.07	0.02	82.70 ± 0.10	0.03	82.84 ± 0.24	0.19
M. caprae (n = 4)	83.24 ± 0.14	0.02	83.11 ± 0.16	0.02	82.97 ± 0.12	0.03	83.11 ± 0.26	0.15
M. microti (n = 6)	83.29 ± 0.09	0.02	83.19 ± 0.09	0.03	83.00 ± 0.10	0.04	83.14 ± 0.24	0.16

Mean values and standard deviation of melting temperatures (T) of a randomly chosen subset of clinical specimen are listed. Corresponding coefficients of variation (CV) in % are indicated for each MTBC species tested.

Table A5

Raw data set and statistical parameters generated from the intra‐ and interassay variability of high‐resolution melting (HRM) assay 2 using a randomly chosen subset of 18 clinical specimens

Isolate			Run 1						Run 2						Run 3						Inter‐Assay
MTBC member	Sample no	Origin	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Ct	Tm	Mean Tm	SD	CV%	Intra‐assay CV%	Mean Tm	SD	CV%	Inter‐assay CV%
M. microti	ATCC 19422		21.28	83.30					21.82	83.25					21.25	83.08					83.21	0.12	0.14
M. bovis BCG Pasteur	ATCC 35734		25.78	82.82					25.33	82.73					25.23	82.62					82.72	0.10	0.12
M. tuberculosis	H37RV		22.87	83.85					22.65	83.67					22.62	83.58					83.70	0.14	0.16
M. tuberculosis	15‐961 2A	Elephant	29.98	83.58	83.59	0.01	0.01		29.48	83.50	83.53	0.03	0.03		29.16	83.47	83.44	0.02	0.03		83.52	0.08	0.09
			29.90	83.60					29.72	83.55					29.33	83.43
			30.23	83.60					29.32	83.53					29.19	83.43
M. tuberculosis	19‐691 1A + B	Elephant	24.51	83.77	83.81	0.03	0.04	0.03	24.78	83.68	83.68	0.03	0.03	0.03	24.44	83.60	83.57	0.03	0.03	0.03	83.68	0.12	0.14	0.12
			24.84	83.82					24.64	83.70					24.49	83.55
			24.63	83.83					24.78	83.65					24.78	83.55
Microsoft. caprae	22971	Cow	31.58	83.20	83.22	0.02	0.02		31.45	83.13	83.13	0.02	0.02		31.21	83.08	82.99	0.08	0.09		83.11	0.11	0.14
			31.37	83.23					30.75	83.12					31.27	82.95
			31.39	83.22					30.96	83.15					31.04	82.95
M. caprae	22914	Cow	29.37	83.17	83.16	0.01	0.01		28.86	83.00	83.02	0.03	0.03		28.57	82.88	82.88	0.02	0.02		83.02	0.14	0.17
			29.04	83.15					28.59	83.05					28.93	82.87
			28.91	83.17					28.54	83.00					28.40	82.90
M. caprae	22848	Cow	19.00	83.35	83.34	0.04	0.04		19.60	83.25	83.26	0.01	0.01		19.69	83.08	83.07	0.01	0.01		83.22	0.14	0.16
			19.33	83.37					19.27	83.27					19.58	83.07
			19.85	83.30					19.16	83.25					19.57	83.07
M. caprae	13‐162	Cow	32.47	83.10	83.11	0.02	0.02	0.02	31.56	82.98	82.98	0.03	0.03	0.02	31.10	82.87	82.87	0.02	0.02	0.03	82.98	0.12	0.15	0.15
			33.03	83.10					31.89	82.95					32.23	82.85
			31.85	83.13					31.93	83.00					31.75	82.88
M. bovis	20593	Cow	14.57	82.90	82.98	0.07	0.08		14.31	82.80	82.79	0.01	0.01		14.18	82.68	82.69	0.01	0.01		82.82	0.15	0.18
			14.67	83.02					14.32	82.80					14.27	82.70
			14.58	83.02					14.23	82.78					14.81	82.70
M. bovis	20594	Cow	12.65	83.07	83.02	0.04	0.05		12.54	82.85	82.85	0.03	0.03		11.91	82.80	82.78	0.02	0.02		82.89	0.12	0.15
			12.60	83.00					12.86	82.83					11.48	82.77
			12.49	83.00					12.58	82.88					11.69	82.77
M. bovis	20609	Cow	22.46	82.87	82.89	0.03	0.04		22.84	82.78	82.76	0.02	0.02		22.50	82.65	82.63	0.03	0.03		82.76	0.13	0.16
			22.49	82.93					22.65	82.75					22.62	82.65
			22.29	82.88					22.45	82.75					22.55	82.60
M. bovis	20608	Cow	13.54	83.03	83.03	0.02	0.02		13.30	82.85	82.84	0.01	0.01		13.11	82.68	82.70	0.02	0.02		82.86	0.17	0.20
			13.19	83.02					13.46	82.85					13.29	82.72
			13.31	83.05					13.16	82.83					13.13	82.70
M. bovis	20596	Cow	12.86	83.02	83.05	0.03	0.03		12.81	82.83	82.83	0.00	0.00		13.02	82.65	82.68	0.04	0.04		82.85	0.18	0.22
			12.93	83.05					12.82	82.83					12.97	82.68
			13.10	83.07					12.84	82.83					12.99	82.72
M. bovis	20600	Cow	13.19	83.00	83.00	0.00	0.00	0.04	12.90	82.82	82.83	0.02	0.02	0.02	13.18	82.65	82.66	0.04	0.05	0.03	82.83	0.17	0.21	0.19
			13.17	83.00					13.08	82.82					12.77	82.62
			13.07	83.00					13.06	82.85					12.95	82.70
M. microti	22928	Cat	21.04	83.28	83.32	0.04	0.04		21.00	83.22	83.22	0.01	0.01		20.82	83.03	83.04	0.01	0.01		83.20	0.14	0.17
			21.25	83.33					21.16	83.22					20.88	83.05
			21.00	83.35					20.93	83.23					20.82	83.05
M. microti	15‐342	Alpaca	15.26	83.35	83.36	0.01	0.01		14.98	83.27	83.24	0.02	0.03		15.10	83.03	83.06	0.02	0.03		83.22	0.15	0.18
			15.15	83.35					14.60	83.23					14.91	83.07
			15.19	83.37					15.04	83.23					15.15	83.07
M. microti	17‐1084	Cat	22.93	83.27	83.29	0.02	0.02		22.99	83.20	83.19	0.02	0.02		23.08	83.05	83.07	0.03	0.03		83.18	0.11	0.13
			22.82	83.30					22.95	83.20					22.87	83.05
			22.90	83.30					23.20	83.17					23.15	83.10
M. microti	15‐1955	Cat	25.48	83.23	83.23	0.02	0.02		25.79	83.10	83.12	0.04	0.05		26.19	83.00	82.99	0.04	0.04		83.12	0.12	0.15
			25.21	83.22					25.61	83.10					25.74	82.95
			25.37	83.25					25.65	83.17					26.35	83.02
M. microti	14‐690	Alpaca	20.03	83.32	83.32	0.00	0.00		20.00	83.20	83.22	0.02	0.02		20.13	83.05	83.03	0.03	0.03		83.19	0.15	0.17
			19.84	83.32					19.90	83.23					19.94	83.05
			19.90	83.32					20.01	83.23					19.98	83.00
M. microti	17‐549	Llama	30.79	83.22	83.23	0.04	0.05	0.02	30.07	83.10	83.13	0.03	0.03	0.03	30.24	82.90	82.96	0.05	0.06	0.04	83.11	0.14	0.17	0.16
			30.23	83.20					30.62	83.15					30.03	83.00
			30.78	83.28					30.20	83.13					29.89	82.98

High‐resolution melting assay 3 showed intra‐assay CVs between 0.01% and 0.02% and an interassay CV of 0.04% (Table 6, Table A4).

Table 6

Intra‐ and interassay variability of HRM assay 3 of clinical specimens

	Run 1		Run 2		Run 3		Interassay
	Tm	CV%	Tm	CV%	Tm	CV%	Tm	CV%
M. tuberculosis H37Rv	86.82		86.82		86.85		86.83	0.02
M. microti ATCC 19422	86.83		86.83		86.83		86.83	0
M. bovis BCG Pasteur ATCC 35734	86.45		86.42		86.47		86.41	0.03
M. bovis (n = 7)	86.86 ± 0.04	0.01	86.84 ± 0.07	0.02	86.88 ± 0.08	0.02	86.86 ± 0.09	0.04

Mean values and standard deviation of melting temperatures (T) of a randomly chosen subset of clinical specimen are listed. Corresponding coefficients of variation (CV) in % are indicated for each MTBC species tested.

Forty‐one NTM and N. paucivorans, E. coli, and S. suis were tested for specificity of HRM assay 2 resulting in no melting curves or melting curves with entirely different T in respect to T deriving from samples of MTBC. In HRM assay 3, some NTM showed similar curves and T values. Therefore, HRM assay 3 is recommended to be applied only after identification of M. bovis or M. bovis BCG.

For HRM assay 2, efficiencies of the qPCR were 94% for M. microti, 97% for M. bovis, 112% for M. caprae, and 91% for H37Rv (Figure A1). In HRM assay 3, efficiencies were determined to be 89% for M. bovis BCG and 103% for M. bovis (Figure A2).

Figure A1

qPCR amplification curves and standard curves of the serial dilution using (a) M. tuberculosis H37Rv; (b) M. bovis BCG Pasteur ATCC 35734; (c) M. microti ATCC 19422; and (d) M. caprae ZH 22914 for high‐resolution melting (HRM) assay 2

Figure A2

qPCR amplification curves and standard curves of the serial dilution using (a) M. bovis BCG Pasteur ATCC 35734 and (b) M. bovis for high‐resolution melting (HRM) assay 3

High‐resolution melting assay 2 showed a LOD for the lowest dilution of which the acceptance criteria (standard deviation < 0.5 and Ct value < 38) were complied with 10 GE, corresponding to 50 fg of template DNA in the qPCR, for M. tuberculosis, M. caprae, M. microti, and M. bovis (Table 7). The LOD of HRM assay 3 was 100 GE for both M. bovis and M. bovis BCG (Table 8).

Table 7

Limit of detection (LOD) of the qPCR of HRM assay 2

MTBC member	Genome equivalents	Ct	SD
M. tuberculosis H37Rv	1,000,000	13.37	0.01
	100,000	16.49	0.28
	10,000	19.99	0.08
	1,000	23.96	0.17
	100	27.46	0.22
	10	30.92	0.40
	1	33.06	1.15
M. bovis BCG pasteur ATCC 35734	1,000,000	12.62	0.02
	100,000	16.02	0.03
	10,000	19.69	0.24
	1,000	23.82	0.19
	100	26.83	0.24
	10	30.11	0.43
	1	32.6	0.61
M. microti ATCC 19422	1,000,000	13.68	0.05
	100,000	17.20	0.13
	10,000	20.80	0.33
	1,000	24.34	0.25
	100	27.77	0.04
	10	31.01	0.49
	1	33.19	0.52
M. caprae ZH 22914	1,000,000	17.96	0.01
	100,000	21.31	0.07
	10,000	25.02	0.06
	1,000	28.28	0.07
	100	31.79	0.29
	10	32.42	0.38
	1	34.09	0.98

Determination of Ct values and its standard deviation (SD) of 3 replicates for a dilution series ranging from 1 to 1,000,000 genome equivalents using reference strains M. tuberculosis H37Rv, M. bovis Pasteur ATCC 35734, M. microti ATCC 19422, and the clinical specimen M. caprae ZH 22914. Bold represents the determined LOD for the lowest dilution of which the acceptance criteria (standard deviation < 0.5 and Ct value < 38) were fulfilled.

Table 8

Limit of detection (LOD) of the qPCR of HRM assay 3

MTBC member	Genome equivalents	Ct	SD
M. bovis ZH 20655	1,000,000	19.19	0.07
	100,000	23.57	0.04
	10,000	28.41	0.31
	1,000	32.45	0.39
	100	36.37	0.10
	10	37.46	0.80
	1	37.67	0.28
M. bovis BCG pasteur ATCC 35734	1,000,000	16.41	0.14
	100,000	20.72	0.18
	10,000	25.06	0.31
	1,000	29.34	0.20
	100	32.83	0.24
	10	35.57	1.84
	1	37.00	2.10

Determination of Ct values and its standard deviation (SD) of 3 replicates for a dilution series ranging from 1 to 1,000,000 genome equivalents using reference strains M. bovis BCG Pasteur ATCC 35734 and the clinical specimen M. bovis ZH 20665. Bold represents the determined LOD for the lowest dilution of which the acceptance criteria (standard deviation < 0.5 and Ct value < 38) were fulfilled.

DISCUSSION

In the present study, the establishment of a three‐reaction HRM paradigm in the form of three HRM assays is described, which can rapidly differentiate the main species of MTBC. M. microti, M. tuberculosis, M. caprae, M. bovis, and M. bovis BCG were demonstrated to undoubtedly and consistently be differentiated from each other by distinctive difference plots (Figures 2c, 3c, and 4c). Based on the recently published suggestion of revising the Hain GenoType MTBC test interpretation, proposing eight possible binding patterns including novel ecotypes/subtypes (Loiseau et al., 2019), the three‐reaction HRM paradigm can potentially reveal the same eight different groups. However, the Hain test is only validated for cultured samples and remains a very costly and time‐consuming approach comprising 13 different probes.

The clear advantage of the developed three‐reaction HRM paradigm approach compared with previous studies (Costa, Amaro, et al., 2014; Costa, Botelho, Couto, Viveiros, & Inácio, 2014; Halse et al., 2011; Pinsky & Banaei, 2008; Pounder et al., 2010; Reddington et al., 2012) is the achievement of an inexpensive, rapid as well as easy to use single‐plex method, which can be used for cultured samples as well as for clinical samples. In approximately 2 hr, a sample can be identified as member of MTBC and assigned to the correct species, which is a benefit comparing to methods, which are based on time‐consuming procedures (Kamerbeek et al., 1997) or require cultured samples (Kasai et al., 2000; Niemann et al., 2000). An additional advantage of this three‐reaction HRM paradigm aiming at different loci lays in the fact that it is an adaptive approach, which opens the possibility for individual combinations of primer pairs depending on the question raised. In case of a probable detection of M. orygis strains and therefore a desired differentiation from M. africanum, there is potential to design a further primer pair, which would allow detecting a mutation at codon 329 of gyrB (Huard et al., 2006) representing a unique SNP in M. orygis.

The only drawback of the described HRM approach compared to the Hain test is the inability to differentiate three strains originating from China (Orloski et al., 2018) (M. caprae/M. bovis ecotype II) from M. caprae. To resolve this problem, it would be possible to expand the assay by designing a novel primer pair covering a SNP at base pair position 1,310 of gyrB and thus reliably identifying M. caprae/M. bovis ecotype II. Since M. caprae/M. bovis ecotype II is a rare type of strains, it can be neglected in routine diagnostic laboratories, where it is essential to rely on simply performable assays. Moreover, an additional differentiation of M. caprae/M. bovis ecotype II from M. caprae has no advantage in respect to the choice of antibiotic treatment, because both strain types are not intrinsically resistant to PZA.

By a stepwise combination of three independent HRM assays, it is possible to differentiate the species of MTBC firstly into three groups (M. microti/M. canettii (rare subtype), M. tuberculosis/M. africanum/M. orygis/M. pinnipedii/Clade A1, and M. caprae/M. bovis/M. bovis BCG/rare M. caprae/M. bovis ecotypes), secondly into six groups (M. microti, M. tuberculosis/M. canettii, M. canettii (rare subtype), M. africanum/M. orygis/M. pinnipedi/Clade A1, M. caprae/rare M. caprae/M. bovis ecotypes, and M. bovis/M. bovis BCG), and finally on gyrA, M. bovis BCG, M. bovis, and rare M. caprae/M. bovis ecotype I can further be separated leading to a clear differentiation into the main human‐ and veterinary‐associated MTBC species (Figure 5). The three HRM assays can be performed either consecutively or in parallel since the qPCR conditions are equal. The interpretation of the three‐reaction paradigm is straightforward and simple to achieve. For routine laboratories, a simple combination of HRM assays 1 and 2 will lead to a rapid detection and differentiation of the most significant agents of tuberculosis appearing worldwide.

Figure 5

HRM assay 1 (Landolt et al., 2019) allows the distinction between M. tuberculosis/M. canettii/M. africanum/M. orygis/M. pinnipedii/Clade A1, M. microti/M. canettii (rare subtype), and M. bovis/M. bovis BCG/M. caprae/rare M. caprae/M. bovis ecotypes. A combination of HRM assays 1 and 2 is leading to six groups (M. tuberculosis/M. canettii, M. africanum/M. orygis/M. pinnipedii/Clade A1, M. microti, M. canettii (rare subtype), M. caprae/rare M. caprae/M. bovis ecotypes I and II, and M. bovis/M. bovis BCG). By performing HRM assay 3, M. bovis, M. bovis BCG, and rare M. caprae/M. bovis ecotype I can further be distinguished. 1. M. africanum not distinguishable in gyrA and gyrB from M. orygis, M. pinnipedii, Clade A1 (Dassie bacillus, M. mungi, Chimpanzee bacillus, M. suricattae) (Brites et al., 2018). 2. Frequent subtype, intrinsic pyrazinamide (PZA) resistance. 3. Rare ecotypes, no intrinsic PZA resistance. 4. Rare subtype, highly recombinogenic

T ranges deriving from HRM assays are partially overlapping (Tables 3, 4, 5, 6). Therefore, based solely on T it is not possible to clearly differentiate the members of MTBC. However, by transforming melting curves into normalized and difference plots using algorithms of the Rotor‐Gene Q Software 2.3.1 (Qiagen), the members of MTBC can be separated into distinct groups (Figures 2, 3, 4). Thereby, the species‐specific melting profiles showed an explicit behavior. The presented HRM assays identified MTBC‐positive cultured isolates in accordance with the results of the GenoType MTBC test.

The clinical samples were in agreement of 96% with respect to the GenoType MTBC test. One sample (4%) deriving from an alpaca, however, showed a nonspecific melting curve preventing a correct assessment of the sample to MTBC applying the developed HRM assays. A possible explanation of this interference of the melting curve could be correlated to the inhibitory substances of the complex sample mixture originating from different tissues including lymph node, lung, heart, liver, and cervical vertebra. Investigations to unravel this finding were not successful yet. Although all remaining 22 clinical samples deriving from lymph nodes, lung, or liver tissues showed an unambiguous and correct result (Table A1), it is important to mention that samples containing very little amount of target DNA material (as obtained in two tested M. caprae samples and one M. microti clinical sample) display a weak fluorescence signal (Figure 4a) and therefore may lead to irregular shapes of the normalized—(Figure 4b) and difference—(Figure 4c) plots in comparison with all other samples enclosing high amounts of MTBC DNA. Such irregular patterns of normalized and difference plots visualize the detection limit of a HRM assay when testing directly extracted clinical samples.

Summarizing the obtained data, both HRM assays 2 and 3 showed a very good reproducibility with small variations of T, which is demonstrated in very low values of intra‐ and interassay CVs. Since HRM assays 1 and 2 are proven to be 100% specific, they can be used unambiguously for the identification and differentiation of MTBC. Moreover, the assays yielded PCR efficiencies of more than 91% and 89%, respectively. The sensitivity of HRM assay 2 showing a LOD of 10 GE is adequate. The LOD of HRM assay 3 is slightly higher with a measured LOD of 100 GE. However, since the tested collective of samples did not cover species of all eight genotype groups, it is suggested to further evaluate the assay by testing a more extensive collection of isolates.

CONCLUSION

The developed three‐reaction HRM paradigm is a quick, sensitive, and specific assay for differentiation of MTBC between the main species highly relevant in human and veterinary diagnostics namely M. tuberculosis/M. canettii, M. canettii (rare subtype), M. africanum/M. orygis/M. pinnipedii/Clade A1, M. microti, M. caprae/rare M. caprae/M. bovis ecotype II, M. bovis, M. bovis BCG, and rare M. caprae/M. bovis ecotype I extracted from clinical samples and from isolates. Several months of cultivation time may be saved by using these potent HRM assays. Since most species within MTBC are implicated in human infections (Huard et al., 2006), it is of advantage to have early knowledge of transmission of tuberculosis for consequently choosing an appropriate drug therapy for humans or a proper eradication strategy when dealing with veterinary samples. Tuberculosis surveillance policies and public health management depend on powerful and affordable diagnostic tools such as this paradigm of a three‐reaction HRM assay, which could be easily implemented in laboratories worldwide.

CONFLICT OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

PL and SS conceptualized, drafted, and investigated the data; involved in formal analysis; and wrote the original manuscript. MJAS edited and provided sequencing data. RS and SS wrote, reviewed, and edited the manuscript.

ETHICS STATEMENT

The recommendations of Swiss federal regulations (TSV 916.401 and VSFK 817.190) were followed. The animal samples were analyzed in the context of a monitoring program of lymph nodes aiming at an early recognition of bovine tuberculosis and NTM infections. No animals were killed for the purposes of this research project, and no ethical approval was required.