Evaluation of Melting Curve Analysis for Screening the Most Prevalent Mutations in Topoisomerase Genes from Streptococcus pneumoniae

In a recently published article, Fukushima et al. (3) described a novel PCR-melting curve analysis (PCR-MCA) method for rapid screening of the most prevalent point mutations related to fluoroquinolone resistance in Streptococcus pneumoniae. In their study, they included 22 levofloxacin (LVX)-resistant strains and 50 LVX-susceptible isolates. None of the isolates in that study were shown to harbor quinolone resistance-determining region (QRDR) silent mutations, and the melting peaks obtained corresponded to the wild-type genotypes or the missense point mutations in the studied codons (gyrA codons 81 and 85 and parC codons 79 and 83).

Using the same approach, our laboratory analyzed 175 S. pneumoniae clinical isolates with ciprofloxacin MICs of ≥0.5 μg/ml (range, 0.5 to 128 μg/ml); 23 were LVX resistant (MIC ≥ 8 μg/ml) and 12 moxifloxacin resistant (MIC ≥ 4 μg/ml). An R6 wild-type strain and derived mutants were used as controls: parC mutants R111 (S79F), R119 (D83Y), and R112 (S79F D83Y) and gyrA mutants R123 (S81F), R125 (E85K), and R124 (S81F E85K). All of the clinical isolates were collected from the Toronto Invasive Bacterial Diseases Network at Mount Sinai Hospital, Toronto, Ontario, Canada. Real-time PCR was performed as described by Fukushima et al. (3), and all gyrA and parC QRDR amplicons were sequenced using an ABI Prism 3100 genetic analyzer (Applied Biosystems, CA).

All missense mutations in parC gene codons 79 (n = 61) and 83 (n = 15) and gyrA gene codons 81 (n = 21) and 85 (n = 1) which were confirmed by sequencing were detected by PCR-MCA. Silent mutations (T→C) associated with the region where the sensor probes bind were detected in 17 clinical isolates by sequencing. Missense point mutations affecting gyrA codon 80 (A→C, D-80-A) and parC codon 78 (G→A, D-78-N) were found in the other two isolates (Table 1). In all of these 19 cases, intermediate melting temperatures (T_m) were observed, with a mean T_m of 49.47°C for parC codon 83 (16 isolates; 0.87% coefficient of variation), 45.35°C for gyrA codon 81 (2 isolates; 0.68% coefficient of variation), 49.09°C for parC codon 79 (1 isolate), and 43.73°C for gyrA codon 85 (1 isolate).

TABLE 1.

Intermediate T_m observed for different S. pneumoniae clinical isolates^a

Isolate	MIC (μg/ml)b			parC codon 79 (sensor, nt 226-243)c		parC codon 83 (sensor, nt 240-257)c		Sequencinge	gyrA codon 81 (sensor, nt 235-251)c		gyrA codon 85 (sensor, nt 244-261)c		Sequencinge
	CIP	LVX	MXF	Tm (°C)	MCA phenotyped	Tm (°C)	MCA phenotyped		Tm (°C)	MCA phenotyped	Tm (°C)	MCA phenotyped
29	8	2	0.25	45.07	M	49.17	I	236(C→T; S-79-F); 249(T→C)	47.64	wt	49.19	wt	WT
32	4	2	0.25	55.39	wt	48.84	I	WT; 249(T→C)	47.99	wt	49.21	wt	WT
100	8	2	0.25	45.75	M	48.98	I	236(C→T; S-79-F); 249(T→C)	47.63	wt	49.21	wt	WT
227	1	1	0.25	55.62	wt	49.37	I	WT; 249(T→C)	47.34	wt	49.39	wt	WT
228	1	1	0.25	55.59	wt	49.89	I	WT; 249(T→C)	47.67	wt	49.41	wt	WT
244	2	2	0.125	49.09	I	52.62	WT	232(G-A; D-78-N)	48.07	wt	49.32	wt	WT
264	4	2	0.25	44.63	M	48.97	I	236(C→T; S-79-F); 249(T→C)	48.37	wt	48.96	wt	WT
295	1	1	0.25	56.98	wt	49.01	I	WT; 249(T→C)	48.09	wt	49.46	wt	WT
296	2	2	0.25	55.99	wt	49.23	I	WT; 249(T→C)	48.56	wt	49.39	wt	WT
299	1	1	0.25	55.54	wt	49.89	I	WT; 249(T→C)	47.85	wt	48.90	wt	WT
327	8	4	1	44.28	M	52.7	WT	S-79-F, K-137-N	45.04	I	49.62	wt	239(A→C; D-80-A)
392	1	1	0.25	55.23	wt	52.6	WT	K-137-N	45.66	I	43.73	I	249(T→C); 340(A→G; S-114-G)
430	32	16	2	44.79	M	49.84	I	236(C→A; S-79-Y); 249(T→C)	41.69	M	49.14	wt	S-81-Y
497	2	2	0.25	55.01	wt	49.93	I	WT; 249(T→C)	47.68	wt	48.96	wt	WT
506	2	2	0.25	55.04	wt	49.92	I	WT; 249(T→C)	47.88	wt	49.33	wt	WT
643	64	16	4	44.21	M	49.82	I	236(C→T; S-79-F); 249(T→C)	41.19	M	49.00	wt	S-81-F
665	2	2	0.25	55.00	wt	49.01	I	WT; 249(T→C)	47.82	wt	49.23	wt	WT
686	4	2	0.25	43.62	M	49.82	I	236(C→T; S-79-F); 249(T→C)	46.91	wt	49.08	wt	WT
690	2	1	0.25	54.82	wt	49.87	I	WT; 249(T→C)	47.44	wt	49.20	wt	WT

^aThe wild-type and missense mutant mean T_m for the four codons studied were, respectively, 55.52°C and 44.95°C for parC codon 79, 52.68°C and 44.41°C for parC codon 83, 47.52°C and 41.85°C for gyrA codon 81, and 47.17°C and 42.30°C for gyrA codon 85. Data in bold indicate the intermediate T_m observed and the point mutations implicated.

^bCIP, ciprofloxacin; MXF, moxifloxacin.

^cThe nucleotides (nt) of the QRDRs covered by the sensor probes are indicated.

^dM, mutant; I, intermediate T_m; wt, wild type.

^eData are presented as follows: nucleotide position (mutation; encoded amino acid change). WT, no mutations in the sequence covered by the sensor probes.

MCA has been shown to be a simple technique that can be performed quickly for the screening of common QRDR mutations to the codon level. Detection of first-step mutations in these four specific codons could prevent clinical treatment failures (1, 2, 4). However, the detection of isolates with intermediate T_m could produce a misinterpretation of the results and indicates that these isolates should be further characterized.