A multisite SNP genotyping and macrolide susceptibility gene method for Mycoplasma pneumoniae based on MALDI-TOF MS.

In this study, a multisite SNP genotyping and macrolide (ML) susceptibility gene test method for Mycoplasma pneumoniae (M. pneumoniae) was developed based on matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). The detection limit of this method for nucleic acids was 102 -103 copies/reaction. Six SNP site-based genotyping and 3 ML susceptibility sites could be detected simultaneously based on multiplex PCR and mass probe. Using the method constructed in this study, 141 Chinese clinical isolates were divided into 8 SNP types. All the SNP test results for the ML susceptibility gene were in line with those of the 23S rRNA sequencing results. With this method, the multisite SNP genotyping and ML susceptibility determination of M. pneumoniae can be completed simultaneously in one test, which greatly reduces the workload and cost, improves the genotyping ability of M. pneumoniae and deserves clinical application.

Introduction

Mycoplasma pneumoniae (M. pneumoniae), an important pathogenic bacterium of respiratory tract infection in children and adults, is responsible for approximately 10%–30% of community-acquired pneumonia (Jacobs et al., 2015; Loens et al., 2010). There has been epidemic spread of M. pneumoniae in many European and Asian countries since 2010 (Blystad et al., 2012; Chalker et al., 2011; Gadsby et al., 2012; Lenglet et al., 2012; Linde et al., 2012; Uldum et al., 2012). Generally, the prognosis of M. pneumoniae is good, with only a few pediatric cases progressing to refractory Mycoplasma pneumoniae pneumonia (RMPP) or severe Mycoplasma pneumoniae pneumonia (SMPP), resulting in severe complications and poor prognosis (Liu et al., 2018; Okumura et al., 2019). To date, macrolides (MLs) have been recommended as the first-line agent for treating clinical infection of M. pneumoniae infections, especially in children (Waites et al., 2009, 2017).

In the past two decades, the proportion of ML susceptibility in M. pneumoniae (MRMP) has shown an increasing trend, especially in Japan and the Republic of Korea (Hong et al., 2013; Kawai et al., 2013). In mainland China, the ML susceptibility ratio of M. pneumoniae is still high to some extent (Cao et al., 2010; Liu et al., 2009; Qu et al., 2013; Xin et al., 2009;Zhao et al., 2012).Studies have shown that the ML resistance mechanism of M. pneumoniae is very fixed and only closely related to 23S rRNA point mutations. The ML susceptibility of M. pneumoniae is closely related to the 23S rRNA mutation, among which the mutations at the 2063 and 2064 positions are correlated with high ML susceptibility (>64 μg/mL for erythromycin) and the mutation at the 2617 position is correlated with low ML susceptibility (Pereyre et al., 2016). Therefore, gene detection based on these mutation sites may serve as an alternative to antimicrobial susceptibility testing of M. pneumonia in vitro. To date, the most common methods of detecting ML resistant of M. pneumoniae strains include real-time PCR-HRM, PCR-RFLP, and pyrosequencing assays (Liu et al., 2014; Matsuoka et al., 2004; Nummi et al., 2015; Peuchant et al., 2009; Spuesens et al., 2010; Wolff et al., 2008).

Genotyping is an important method for studying the molecular biology and epidemiologic features of M. pneumoniae. There may be a close relationship between the transformation of M. pneumoniae genotypes in the population and the epidemics of M. pneumoniae, and genotyping is also an important technique for tracing the origin of M. pneumoniae local outbreaks. The M. pneumoniae genome is highly conserved, with a sequence similarity of up to 99% among strains. Based on single nucleotide polymorphisms (SNPs) and indels, two main subgroups of M. pneumoniae, genotype I and genotype II, can be clearly differentiated (Lluch-Senar et al., 2015; Xiao et al., 2015). Currently, the most common genotyping techniques for these two genotypes are based mainly on PCR-RFLP genotyping of the p1 gene (Dorigo-Zetsma et al., 2000), VNTR of the pl gene, and real-time PCR of the MPN459/MPNA5864 gene (Zhao et al., 2015). These techniques are easy to perform, but the discrimination ability is not adequate due to the presence of a single target site. Specifically, multiple-locus variable-number tandem-repeat analysis (MLVA) (Degrange et al., 2009) and multilocus sequence typing (MLST) (Brown et al., 2015) contribute to the genotyping of M. pneumoniae. However, these genotyping procedures are time-consuming and technically demanding, which hampers their applications. Both genotyping analysis and ML susceptibility nucleic acid testing for M. pneumoniae involve complicated procedures, and it is necessary to develop new methods that are rapid and inexpensive.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been utilized for detecting SNPs in recent decades. Multiplex PCR is used to amplify the genes containing the targets of SNPs. Subsequently, an extension mass probe has been utilized for the extension of SNP sites. Finally, MALDI-TOF MS is performed to identify the mass-to-charge ratio (m/z) of extended mass probes. Recently, MALDI-TOF MS has been commonly used in microbiological detection and analysis, and is considered a new gold standard for the identification of many kinds of microorganisms (Jang and Kim, 2018;Kostrzewa, 2018;Rahi et al., 2016). To date, SNP genotyping based on MALDI-TOF MS has been frequently conducted for the screening of hand-foot-and-mouth disease virus, hepatitis B virus (HBV) detection, and the human papillomavirus (HPV) genotyping and serves as an effective alternative to conventional techniques (Peng et al., 2013a, 2013b;Sjoholm et al., 2008). SNP analysis based on MALDI-TOF MS features high throughput, rapid detection, and simultaneous detection of multiple targets. Therefore, it is one of the most promising biomolecular techniques.

In this study, multiplex PCR coupled with MALDI-TOF MS (PCR-MALDI-TOF MS) was used to genotype and detect the antimicrobial susceptibility of M. pneumoniae.

Results

MALDI-TOF MS-based method establishment and optimization

Multiplex PCR coupled with MALDI-TOF MS assay was utilized to automate the detection of the nine SNPs and avoid sequencing the gene fragments. To avoid the formation of a dimer between the probes at the 23S rRNA2063 and 23S rRNA2064 sites, the quality probe at the 2064 site was designed at the reverse complementary strand. Therefore, the SNP of this site was the basis of the complementary strand (Table 1, Figure 1). MALDI-TOF MS-based amplification and identification were performed on 10 epidemiologically unrelated M. pneumoniae strains. Six SNP genotyping sites and three ML susceptibility gene sites were detected, and no MS peak was observed in the blank control. There was complete consistency between the MS SNP results of three ML susceptibility sites and the sequencing results of the strains. Among the 10 strains, there were 6 SNP types (0, 1, 3, 11, 15 and 32). Six odd number SNP types were classified into genotype I, and 4 even number SNP types were classified into genotype II (Table 2) . For example, there were 9 MS peaks for M129 in Figure 1A, representing 6 SNP typing mass probes and 3 ML susceptibility mass probes. The dotted line of the same color represents the mass spectrum peak for various mass probes extending the same base pairs in the SNP sites. The peak value represents the m/z of the MPE after extension. After multiplex PCR amplification and MPE probes extension for M129, the peak of the spectrum is shown in Figure 1B. The red arrow represents the quality value of the nine MPE probes after SNP site extension. The optimized concentration of each extension primer in the primer mixture was 6.9 μM (MPN1141461), 7.1μM (MPN3721112), 7.7μM (23S rRNA2063), 8.0μM (MPN21347), 8.6μM (MPN2801641), 9.2μM (23S rRNA2617), 9.5μM (MPN126470), 10.4μM (MPN262192), and 11.2μM (23S rRNA2064).

Table 1

Primer sequences for M. pneumoniae SNP typing and susceptibility target site amplification and MPE probes for SNP detection

	Target gene	SNP	Multiple-PCR		Mass probe extension
			Forward primer sequence	Reverse primer sequence	Mass probe sequence	Mass probe mass (Da)	Extension call	Extended mass (Da)	Extension call	Extended mass (Da)	Extension call	Extended mass (Da)	Extension call	Extended mass (Da)
GENOTYPING	MPN1141461	C/T	acgttggatgCACCGAGTGTCTTCGTCCTT	acgttggatgGGTTGACCCCACTACCCTTT	CCCTTTTGCGCTAACCG	5097.4	C	5410.4	T	5394.4
	MPN126470	C/T	acgttggatgAAATTCCCCGTAATCCGAGA	acgttggatgGCCTTCTGCTTGTTAGGTAGATG	AGTGATTTAACGTTCCATTTTC	6690.4	C	6963.4	T	7032.4
	MPN21347	G/T	acgttggatgATCAGTCGCTGGTACTGTGG	acgttggatgGCTGTGCTTTTGGCAACTTA	TACTACCACCGCAGTTGTT	5738.8	G	6011.8	T	6035.8
	MPN262192	C/T	acgttggatgACCACTTAGAGCCGGAACAA	acgttggatgTTGCTGAAACTGACCATCCA	GCCCAATTTAGTCTTATTTTGGAT	7323.8	C	7636.8	T	7620.8
	MPN2801641	G/T	acgttggatgCTCAATTAAACGCGGTTTGG	acgttggatgCGGTTGAAATCAACGGGTTA	TTCCAAATGAAGTAAGACAC	6118.0	G	6391.0	T	6415.0
	MPN3721112	G/T	acgttggatgAGTGTTAGCGCGGTTAATGC	acgttggatgTCGTGTTGTGAACCACCACT	AATGACACCGCAAGACA	5181.4	T	5523.4	G	5494.4
ML DETECTION	23S rRNA2063	A/T/C/G	acgttggatgCCAGGTACGGGTGAAGACAC	acgttggatgATTCCACCTTTCGCATCAAC	TTAGGCGCAACGGGACGG	5589.6	G	5902.6	A	5886.6	C	5862.6	T	5931.6
	a23S rRNA2064	A/T/C/G	acgttggatgCCAGGTACGGGTGAAGACAC	acgttggatgATTCCACCTTTCGCATCAAC	AGCTACAGTAAAGCTTCACGGGGTCT	7995.2	T	8337.2	C	8268.2	G	8308.2	A	8292.2
	23S rRNA2617	A/T/C/G	acgttggatgGCTGTTCGCCGATTAAAGAG	acgttggatgGCGCTACAACTGGAGCATAA	CCGTCGTGAGACAGGTTGGTC	6478.2	C	6751.2	G	6791.2	A	6775.2	T	6820.2

The MPE extension probe for the SNP of 23S rRNA2064 was the reverse complementary strand, which detected the complementary bases for the SNP.

Figure 1

MS peak of the MPE probes

(A) MS peaks of 9 MPE probes without extension;

(B) SNP peak of the 9 MPE probes extended with M. pneumoniae M129.

Table 2

Characteristics of the ten M. pneumoniae clinical isolates and strains selected for method establishment

Strain/Isolate	Year of isolation	Susceptibility to macrolides	Sequence 23S rRNA mutation	P1 gene type	MLVA	Geographical origin	SNP typing	ML SNP 23S rRNA mutation	Reference
M129(ATCC29342)	1969	S	None	1	4572	USA	SNP1	None	Himmelreich et al. (1996)
FH(ATCC15531)	1954	S	None	2	3562	Boston, USA	SNP0	None	Xiao et al. (2015)
ICDC SY15-11	2012	S	None	1	4572	Beijng, China	SNP11	None	Zhao et al., 2013b
ICDC P028	2008	S	None	2	3562	Beijng, China	SNP32	None	Zhao et al., 2013b
ICDC 21075	2016	S	None	2	3572	Beijng, China	SNP0	None	Zhao et al. (2019a)
ICDC AH-012	2017	R	A2063G	1	4572	Fuyang, China	SNP11	A2063G	Zhao et al. (2019b)
ICDC SD14-08	2017	R	A2063G	2	3562	Jinan, China	SNP0	A2063G	Zhao et al. (2019b)
ICDC SL066	2018	R	A2064G	1	4572	Jilin, China	SNP3	A2064G	Zhao et al. (2019b)
ICDC JSSZ031	2018	R	A2063G	1	4573	Soochow, China	SNP15	A2063G	Zhao et al. (2019b)
ICDC W081	2019	R	A2063T	1	4573	Weihai, China	SNP15	A2063T	This study

Specificity and detection limit of PCR-MALDI-TOF MS

There was no extension signal on the quality probe for the nucleic acids from the 20 non-M. pneumoniae pathogens, yielding a specificity of 100%. In addition, the nucleic acids of M. pneumoniae M129 strains subjected to 10-fold dilution (100-105copies/μl) were utilized for the validation of LDL, using 1 μL template. In the presence of an M. pneumoniae load of 103-105copies/reaction, all the nine SNP sites could be accurately identified after two tests. In the presence of a load of 102 copies/reaction, 23S rRNA2617 site could be accurately examined only once, while there was no MS peak in the other test. For MPN262, there was no MS peak after two tests, and no signals were noticed in the MS peak in the blank control (Figure 2). The detection limit of the nucleic acid derived from the M. pneumoniae strains was in the range of 102-103copies/reaction.

Figure 2

Detection of 9 SNP sites in the presence of different nucleic acid from M. pneumoniae based on the PCR-MALDI-TOF MS method

The test was performed twice for each concentration. a-g: of a concentration range of 105-100copies/μL and the blank control.

Nucleic acid detection of M. pneumoniae isolates

The genotyping and ML susceptibility of the nucleic acids from 141 M. pneumoniae isolates were evaluated using this method. The genotyping and ML susceptibility results are shown in Table 3. For the ML susceptibility sites, 106 were confirmed with the A2063G mutation, 3 with the A2064T mutation, 1 with the A2063T mutation, and 31 with no mutations. The MS findings were consistent with previous sequencing results and drug sensitivity tests. SNP genotyping indicated that there were eight genotypes for the 141 isolates, namely, SNP0, 1, 3, 11, 15, 26, 32, and 27. The 109 strains of genotype I showed an odd number SNP genotype (1, 3, 11, 15, 17), among which SNP 11, 15, and 27 were the major types. Thirty-two genotype II strains showed an even number SNP genotype (0, 26, 32), among which SNP32 was the major type. There were large variations in the number and proportion of the SNP genotypes in the M. pneumoniae isolates (Table 4).

Table 3

Genotyping and macrolide susceptibility results of M. pneumoniae strains used in this study

Isolates name	City	year of isolation	MLVA type	Genotype	Nucleotide of SNP						SNP1-6 to digital string	SNP type	SNP (odd/even)	23S rRNA SNP
					SNP1	SNP2	SNP3	SNP4	SNP5	SNP6				2063	2064	2617
					aMPN1141461	MPN126470	MPN21347	MPN262192	MPN2801641	MPN3721112				23S rRNA2063	23S rRNA2064	23S rRNA2617
ICDC SD14-02	Jinan	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-07	Jinan	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-09	Jinan	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-26	Jinan	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-50	Jinan	2018	3662	II	C	C	G	C	G	G	000000	0	even	G	A	C
ICDC SD14-58	Jinan	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-61	Jinan	2018	3562	II	C	C	G	C	G	G	000000	0	even	G	A	C
ICDC SD14-69	Jinan	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-77	Jinan	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC SD14-13	Jinan	2017	3562	II	C	T	T	C	T	G	011010	26	even	A	A	C
ICDC SD14-91	Jinan	2018	3562	II	C	T	T	C	T	G	011010	26	even	A	A	C
ICDC SD14-03	Jinan	2017	4572	I	C	C	G	C	T	T	000011	3	odd	A	G	C
ICDC SD14-01	Jinan	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SD14-06	Jinan	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SD14-22	Jinan	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SD14-23	Jinan	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SD14-37	Jinan	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SD14-88	Jinan	2018	4572	I	C	C	T	C	T	T	001011	11	odd	A	A	C
ICDC SD14-45	Jinan	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SD14-14	Jinan	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SD14-25	Jinan	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL002	Jilin	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL008	Jilin	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL011	Jilin	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL019	Jilin	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL033	Jilin	2018	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC SL037	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL044	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL061	Jilin	2018	4573	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL066	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL069	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL075	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL083	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL087	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL091	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL093	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL121	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL134	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL144	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL145	Jilin	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SL181	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL182	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL199	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL200	Jilin	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL251	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL266	Jilin	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SL277	Jilin	2018	4472	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC SL281	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL288	Jilin	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SL309	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC SL331	Jilin	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH001	Beijing	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC BCH077	Beijing	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC BCH336	Beijing	2018	3562	II	C	C	G	C	G	G	000000	0	even	G	A	C
ICDC BCH423	Beijing	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC BCH437	Beijing	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC BCH443	Beijing	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC BCH118	Beijing	2017	3562	II	C	T	T	C	T	G	011010	26	even	A	A	C
ICDC BCH023	Beijing	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC BCH031	Beijing	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC BCH039	Beijing	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC BCH145	Beijing	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC BCH331	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH012	Beijing	2017	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH057	Beijing	2017	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH112	Beijing	2017	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH120	Beijing	2017	4573	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH133	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH178	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH215	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH281	Beijing	2018	4472	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH399	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH451	Beijing	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC BCH008	Beijing	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH016	Beijing	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH019	Beijing	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH022	Beijing	2017	4572	I	C	T	T	C	T	T	011011	27	odd	A	A	C
ICDC BCH090	Beijing	2017	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH264	Beijing	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH271	Beijing	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC BCH412	Beijing	2018	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC U016	Beijing	2008	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC U098	Beijing	2008	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC 12013	Beijing	2012	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC 21055	Beijing	2012	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC 21077	Beijing	2012	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC 21109	Beijing	2013	3562	II	C	C	G	C	G	G	000000	0	even	A	G	C
ICDC B117	Beijing	2013	3562	II	C	C	G	C	G	G	000000	0	even	G	A	C
ICDC P0118	Beijing	2008	3562	II	T	C	G	C	G	G	100000	32	even	A	A	C
ICDC P033	Beijing	2008	3562	II	T	C	G	C	G	G	100000	32	even	A	A	C
ICDC P054	Beijing	2008	3562	II	T	C	G	C	G	G	100000	32	even	A	A	C
ICDC 21010	Beijing	2012	4572	I	C	C	G	C	G	T	000001	1	odd	G	A	C
ICDC SR15-22	Beijing	2015	4472	I	C	C	G	C	G	T	000001	1	odd	G	A	C
ICDC U014	Beijing	2008	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC P396	Beijing	2009	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC 21001	Beijing	2012	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC 12066	Beijing	2013	4572	I	C	C	T	C	T	T	001011	11	odd	T	A	C
ICDC B271	Beijing	2013	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC SY15-77	Beijing	2015	4572	I	C	C	T	T	T	T	001111	15	odd	A	G	C
ICDC P005	Beijing	2008	4572	I	C	C	T	T	T	T	001111	15	odd	A	A	C
ICDC 12075	Beijing	2013	4572	I	C	C	T	T	T	T	001111	15	odd	A	A	C
ICDC B023	Beijing	2013	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC B370	Beijing	2013	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC B385	Beijing	2014	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC P135	Beijing	2008	4573	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC 10577	Beijing	2012	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC 12038	Beijing	2012	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC 21000	Beijing	2012	4573	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC 21114	Beijing	2013	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC B429	Beijing	2014	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC B434	Beijing	2014	4572	I	C	T	T	C	T	T	011011	27	odd	G	A	C
ICDC JSSZ002	Soochow	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC JSSZ012	Soochow	2017	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC JSSZ023	Soochow	2017	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC JSSZ038	Soochow	2017	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC JSSZ007	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ010	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ024	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ028	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ037	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ041	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ046	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ047	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ048	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ051	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ055	Soochow	2017	4573	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ064	Soochow	2017	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ072	Soochow	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ078	Soochow	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC JSSZ033	Soochow	2017	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC JSSZ067	Soochow	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC JSSZ071	Soochow	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC JSSZ073	Soochow	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC JSSZ077	Soochow	2018	4572	I	C	C	T	T	T	T	001111	15	odd	G	A	C
ICDC AH-001	Fuyang	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC AH-014	Fuyang	2018	3562	II	C	C	G	C	G	G	000000	0	even	A	A	C
ICDC AH-007	Fuyang	2018	4572	I	C	C	G	C	T	T	000011	3	odd	G	A	C
ICDC AH-004	Fuyang	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC AH-006	Fuyang	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC AH-008	Fuyang	2018	4472	I	C	C	T	C	T	T	001011	11	odd	G	A	C
ICDC AH-010	Fuyang	2018	4572	I	C	C	T	C	T	T	001011	11	odd	G	A	C

The result information for columns 1–5 in the table comes from reference Zhao et al. (2013a), 2013b, Zhao et al. (2019b).

ATCC29342 (GenBank: U00089.2) was used as reference. The number after MPN is the gene number, and the superscript number indicates the position of SNP in the gene.

Table 4

SNP genotyping of 141 Chinese M. pneumoniae isolates, macrolide susceptibility test results, and mimic genotyping of 79 international M. pneumoniae

	Year	No	Genotype 1					Genotype 2			23S rRNA
			SNP1	SNP3	SNP11	SNP15	SNP27	SNP0	SNP26	SNP32	A2063G	A3064G	A2063T	None
Beijing, China	2008–2012	30	2	2	3	6	7	7	0	3	17	2	1	10
Beijing, China	2017–2018	30	0	0	4	11	8	6	1	0	23	0	0	7
Soochow, China	2017–2018	23	0	2	14	5	0	2	0	0	21	0	0	2
Jinan, China	2017–2018	21	0	1	6	1	2	9	2	0	10	1	0	10
Jilin, China	2017–2018	30	0	1	3	7	19	0	0	0	30	0	0	0
Fuyang, China	2017–2018	7	0	1	4	0	0	2	0	0	5	0	0	2
Total	2008–2018	141	2	7	34	30	36	26	3	3	106	3	1	31
International strains	–	79	14	17	4	7	7	30	0	0	–	–	–	–

Detailed information on each M. pneumoniae strain is shown inTable 3.

Mimic SNP genotyping of international M. pneumoniae strains

There were 6 genotypes, namely, SNP 0,1,3,11,15 and 27, among the genomes of 79 international M. pneumoniae strains with sequences in the NCBI databases. There were 30 strains with SNP 0 genotypes, including AP012303.1, AP017318.1, AP017319.1, CP002077.1, CP010539.1, CP010540.1, CP010546.1, CP010547.1, CP010548.1, CP010549.1, CP010550.1, CP010551.1, CP017327.1, CP017329.1, CP017334.1, CP017335.1, CP017336.1, CP017337.1, CP017338.1, CP017339.1, CP017340.1, CP017341.1, CP017342.1, CP039761.1, CP039772.1, CP039775.1, CP039777.1, CP039781.1, CP039784.1, and LR214945.1. There were 14 SNP 1 genotype including CP003913.2, CP017330.1, CP017343.1, CP020689.1, CP020690.1, CP020691.1, CP020692.1, CP020693.1, CP020710.1, CP020711.1, CP020712.1, CP039787.1, CP039790.1, and U00089.2. There were 17 SNP 3 genotype including CP008895.1, CP010538.1, CP010541.1, CP010542.1, CP010543.1, CP010544.1, CP010545.1, CP017328.1, CP017331.1, CP017332.1, CP017333.1, CP039776.1, CP039779.1, CP039782.1, CP039783.1, CP039785.1, and CP039788.1. There were 4 SNP11 genotypes (i.e. CP013829.1, CP014267.1, CP039773.1, and CP039780.1). There were 7 SNP16 genotypes (CP039762.1, CP039763.1, CP039764.1, CP039765.1, CP039766.1, CP039767.1, and CP039769.1) and 7 SNP27 genotype (CP039768.1, CP039770.1, CP039771.1, CP039774.1, CP039778.1, CP039786.1, and CP039789.1). Among these genotypes, all the odd number SNP genotypes were genotype I, with SNP1 and SNP3 as the major types, and all the even number SNP genotypes were genotype II. Compared with the Chinese strains, the international strains lacked the genotypes SNP26 and SNP32 (Table 4).

Detection results for the nucleic acid positive clinical specimens of M. pneumoniae

Our assay was performed directly on clinical specimens using an increase in the template volume (5 μL) and PCR volume (20 μL). Among the 30 M. pneumoniae nucleic acid positive clinical specimens, 9 SNP sites were detected in 25 isolates (83.3%). The detection limit of the clinical specimens was 5.2×102 copies/reaction. In addition, the genotype and ML susceptibility results in the 25 isolates were consistent with the detection of M. pneumoniae. For the remaining 5 specimens, incomplete SNP profiles were obtained. On this basis, complete data interpretation could not be given.

Discussion

With the increased accessibility to markers through high-throughput whole-genome sequencing (WGS), SNP analysis is increasingly useful for evaluating drug susceptibility, evolution, and molecular epidemiology (Kuglik et al., 2014; Lipworth et al., 2019). To date, the single or small proportion of SNP site detection mainly relies on the first-generation sequencing technique or TaqMAN probe-based real-time PCR detection. For the analysis of multiple SNP sites that are extensively distributed, the WGS technique is the preferred option. However, these techniques are not adequate in the detection of more than tens or dozens of SNP sites. The MS technique facilitates these two SNP-based techniques. Multiplex PCR and the single base pair extension technique have proven to be superior in detecting more than ten SNP sites and are extensively utilized in the genotyping and identification of multiple microorganisms(Jang and Kim, 2018; Kostrzewa, 2018; Peng et al., 2013a, 2013b;Rahi et al., 2016;Sjoholm et al., 2008).

To date, there is still a lack of efficiency regarding genotyping techniques for M. pneumoniae compared with other pathogens. Currently, the most commonly utilized methods are based on the genotyping technique for the two genotypes. These methods are easy to perform, but their application in tracing pathogenic bacteria is hampered due to insufficient discriminatory capacity. In 2009, Degrange et al. established an MLVA genotyping technique using 5 VNTR sites, which divided M. pneumoniae into dozens of MLVA genotypes. However, the classification ability of this method is still limited in practical applications (Benitez et al., 2012;Degrange et al., 2009;Dumke and Jacobs, 2011;Kubota et al., 2015;Pereyre et al., 2013;Qu et al., 2013; Waller et al., 2014;Xue et al., 2014;Yan et al., 2014;Zhao et al., 2013a, 2013b). In 2015, Brown et al. (Brown et al., 2015) established a genotyping technique of MLST. MLST improves the typing ability to a certain extent, but such a technique is not extensively utilized, as it is technically demanding and expensive. In addition, there is a necessity for data uploading. In the same year, Touati et al. developed SNaPshot minisequencing technology based on the single base extension of a specially designed minisequencing primer, which promoted the discrimination of M. pneumoniae. However, the cost was still high, even though it was less than that presented in a previous study (4.2 Euro) (Touati et al., 2015). In this study, 6 SNP genotyping sites, which were derived from 1434 SNPs (Table S1) screened from 20 M. pneumoniae clinical isolates, were utilized as the targets. In addition, all three sites associated with ML susceptibility were included, and then MS-based amplification was established that could simultaneously accomplish M. pneumoniae genotyping of multiple sites and ML susceptibility analysis, which could test 96 samples simultaneously within 6 hr based on PCR amplification and extension (4.5 hr) and MS analysis (0.5 hr). The cost for MS was only 1.0 RMB (approximately 0.1 Euro), which was far less than that reported by Touati (Touati et al., 2015) Meanwhile, our method could provide more data on the ML susceptibility gene based on genotyping. To the best of our knowledge, this method provides the highest throughput for the M. pneumoniae genotyping and ML susceptibility gene testing and involves less time and expenditure than the current methods.

The selected SNP site, MPN3721112 (G1112T), was closely related to the two genotypes of M. pneumoniae, with T and G for genotype I and genotype II, respectively. According to the data interpretation standard, stains with an even number SNP genotype were classified into genotype II, while those with an odd number SNP genotype were classified into genotype I (Table S2). The SNP genotyping of the 141 Chinese isolates showed complete consistency, among which 109 genotype I strains showed an odd number SNP genotype and the 32 genotype II strains showed an even number SNP genotype. Moreover, for the 79 international strains, genome sequencing was consistent with the above rules (Table 3). This indicated that our method could improve the discriminatory capacity and present the genotyping traits of the two conventional genotypes of M. pneumoniae. For the detection of ML susceptibility sites of 23S rRNA2063, 23S rRNA2064 and 23S rRNA20617, 106 isolates showed A2063G, 3 showed A2064G, 1 showenA2063T, and 31 showed no mutations at the 2063/2064 sites. SNP ML susceptibility findings were consistent with the strain sequencing and ML sensitivity test findings (Table 3). The SNP genotyping results indicated that there were 4–7 SNP genotypes in the five selected cities (Table 4). There were differences in the SNP genotypes in different regions between 2017 and 2018. Specifically, SNP27 was the predominant type in Jilin City in northern China, while SNP14 was found in Suzhou in southern China, SNP15 in Beijing in central China, and SNP0 in Jinan in the mid-eastern part of China. The differences in SNP types for the same region in different years were relatively small. For instance, the genotype I M. pneumoniae strains between 2008 and 2012 as well as 2017 and 2018 in Beijing were predominantly of the SNP15 and SNP27 types, while for genotype II, SNP0 was the predominant type. For the 79 international strains, there were 6 SNP genotypes, among which 49 genotype I M. pneumoniae strains showed an odd number SNP type and 30 genotype II M. pneumoniae strains showed an even number SNP type. The international strains with an odd number SNP type were predominantly SNP1 and SNP3, while those of the Chinese strains were SNP11, SNP15, and SNP27, yielding a large variance. This further confirmed the discriminatory capacity of our method.

In addition to genotyping and the ML susceptibility test with pure culture M. pneumoniae nucleic acids, our method could be utilized for the analysis of nucleic acids derived from clinical specimens. The detection limit of the clinical specimens was 5.2×102 copies/reaction, which was relatively consistent with that of nucleic acids isolated from the purified M. pneumoniae strains (102-103copies/reaction). The detection limit was on the same order of magnitude (102 copies/reaction), which was similar to that of the multiplex PCR. The genotyping and ML susceptibility positive rate of the 30 clinical isolates with nucleic acid load after real-time PCR was 83.3%. This implied that our method contributed to the genotyping of the clinical isolates and the ML susceptibility test, which was appropriate for the rapid genotyping of M. pneumoniae with a slow growth and rapid detection of ML susceptibility.

In this study, we developed a multiplex PCR coupled with the MALDI-TOF MS method for the multisite genotyping and ML susceptibility gene testing of M. pneumoniae. The method was easy to perform with a high specificity and a low cost as well as a high throughput. Multisite genotyping and ML susceptibility gene testing could be accomplished simultaneously, with a strong discriminatory capacity for strains of genotype I and genotype II. The analysis of the ML susceptibility site was comprehensive and accurate and deserves clinical application. Meanwhile, the PCR-MALDI-TOF MS technique could meet the demands of various regimens, which may contribute to the microbial identification, genotyping, and drug susceptibility testing.

Limitations of the study

The sensitivity of the method constructed in this study is limited by the detection limit of multiplex PCR technology.

Resource availability

Lead contact

Further information and requests for resources should be directed to and will be fulfilled by the Lead Contact, Prof. Di Xiao, xiaodi@icdc.cn.

Material availability

This study did not generate nor use any new or unique reagents.

Data and code availability

The raw data of MALDI-TOF MS of all the samples used in this study are available on the project homepage at https://dx.doi.org/10.17632/b8pnzp2y6c.1.

Methods

All methods can be found in the accompanying Transparent methods supplemental file.