Importance of a Diverse Isolate Collection When Defining Genotype-Specific Mass Spectra in Staphylococcus aureus

LETTER

Strain typing would be a way to expand the use of matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS). Wolters et al. (1) and Josten et al. (2) have shown the possibility of differentiating clonal lineages in Staphylococcus aureus of human origin using MALDI-TOF fingerprinting. To explore this, a collection of 77 methicillin-resistant Staphylococcus aureus (MRSA) isolates of human origin, representing clonal complex 5 (CC5), CC7, CC8, CC22, CC30, CC45, CC59, CC80, and CC398 deduced by spa typing, was received from the National Reference Laboratory for Staphylococci at Statens Serum Institut, Copenhagen, Denmark. All strains were retrieved on blood agar at 35°C overnight and extracted with formic acid as previously described (1, 2). One microliter of extract was spotted on a target plate, left to dry, and overlaid with 1 μl of α-cyano-4-hydroxycinnamic acid (HCCA) matrix (Bruker Daltonics). Measurements were performed on a Microflex (Bruker Daltonics). Four spectra per isolate were obtained (240 laser shots/spectra) and examined with the focus on the presence or absence of the previously reported CC-specific peaks/peak patterns (1, 2) using the Flexanalysis 3.3 software (Bruker Daltonics). Ten peak patterns (“MALDI MS types”) could be distinguished (Table 1). Five of the CCs reported by Wolters et al. and six of the CCs analyzed by Josten et al. were represented in this collection with the addition of CC7, CC59, and CC80 (Table 1). Wolters et al. reported 13 peaks to differentiate the five CCs (1). Four of these peaks (m/z 3276, 3876, 6612, and 7734) were either present or absent in all spectra in our study and thus did not add anything to the analysis. The peak patterns reported specific to CC8 (presence of m/z 6591) and CC22 (absence of m/z 5032) (1, 2) could be verified, and CC5, CC80, and CC398 were separated from the other CCs by a common peak pattern (Table 1). The latter pattern was reported by Wolters et al. to be CC5 specific, but CC80 and CC398 were not included in their study (1). Compared to Josten et al., we found the peak of m/z 3891 (2) to be both insensitive and unspecific for the detection of CC5, and similarly, we could not repeat the finding of the peak of m/z 4511 in CC398 (2). This peak might be related to the spa type t011, predominating in the Netherlands, and cannot be considered a sensitive peak for detecting CC398 in general. CC45 was distinguished by two peak patterns (Table 1) of which one was new (1). For CC30, two MALDI MS types could be separated from other CCs except for CC59 (Table 1). The peak of m/z 5508, previously reported as CC30 specific (1, 2), was absent in 2/9 CC30 isolates and present in all CC59 isolates. CC7 (one isolate) displayed a unique peak pattern.

TABLE 1.

Separation of 77 MRSA isolates from diverse origins into MALDI MS types

MALDI MS type	n	CC	Peak at m/za
			4511b	4641	5002b	5032b	5419	5437	5508b	5524	6591b
MT001	4	22	0	0	1	0	0	1	0	1	0
MT002	4	22	0	0	0	0	0	1	0	1	0
MT003	9	8	0	0	0	1	0	1	0	1	1
MT004	12	8	0	1	0	1	0	1	0	1	1
MT005	18 + 4 + 2	5 + 80 + 398	0	0	0	1	0	1	0	1	0
MT006	6	45	1	0	0	1	0	1	0	1	0
MT007	2	45	1	1	0	1	0	0	0	0	0
MT008	2	30	1	0	0	1	0	0	0	1	0
MT009	7 + 6	30 + 59	1	0	0	1	1	0	1	0	0
MT010	1	7	0	0	0	1	0	0	0	1	0

^aPeaks listed are based on discriminatory peaks proposed by Wolters et al. (1) and in some cases also proposed by Josten et al. (2). The presence and absence of the peak in this study is indicated by a 1 and 0, respectively.

^bPeaks reported by Josten et al. (2).

With this collection representing nine clonal complexes, we could verify that CC22 and CC8 can be well differentiated from other CCs using MALDI-TOF MS. Although the other CCs could be distinguished in this material, we found discrepancies in the peak patterns between our collection and the previous studies. Our findings are in line with another very recent study (3), and multicenter studies of large number of isolates of both MRSA and methicillin-resistant S. aureus (MSSA) with diverse origin seem necessary to further determine the sensitivity and specificity of CC-specific peak patterns in S. aureus.