Analysis of Phenotypic Variants of the Serogroup C ET-15 Clone of Neisseria meningitidis by Pulsed-Field Gel Electrophoresis

In Canada, two waves of hyperendemic meningococcal diseases were documented in the last 20 years. Both were due to a unique clone of serogroup C Neisseria meningitidis designated by multilocus enzyme electrophoresis as ET-15 (1). When this clone first appeared, it was characterized by the antigenic formula C:2a:P1.5,2 (1). In 2001, antigenic variants of this ET-15 clone characterized as C:2a:P1.7,1 (7) or C:2a:P1.5 (4) emerged to cause outbreaks (8). In this study, we examined whether pulsed-field gel electrophoresis (PFGE) can be used as a discriminatory tool to differentiate between the antigenic variants C:2a:P1.7,1, C:2a:P1.5, and C:2a:P1.5,2 of the ET-15 clone of N. meningitidis.

Eleven serogroup C ET-15 N. meningitidis isolates from invasive meningococcal disease cases were selected for this study. Ten of the 11 serogroup C isolates were identified as serotype 2a. DNA sequencing of the porB gene of the nontypeable isolate identified it as a serotype 2a mutant containing a previously described mutational hot spot (2). There were four different combinations of serosubtype antigens observed for these 11 isolates: 5 with the P1.5 antigen, 3 with the P1.7,1 antigens, 2 with the P1.5,2 antigens, and 1 with the P1.2 antigen. The porA genes of these strains were sequenced, and their PorA variable regions (VRs) are summarized in Fig. 1.

FIG. 1.

Eleven PFGE profiles of SpeI-digested DNA (using Dice coefficient with a position tolerance of 1.5% and optimization of 2.0% on BioNumerics 3.5 software) representing the three Neisseria meningitidis serogroup C ET-15 antigenic variants C:2a:P1.5,2, C:2a:P1.7,1, and C:2a:P1.5. The asterisk indicates that the phenotype of the strain was C:NT:P1.2. Antigens in parentheses were deduced by DNA sequencing of the serotype and serosubtype antigen genes. This serotype 2a strain has a single nonsynonymous point mutation that led to the nonserotypeable phenotype (7).

Multilocus sequence typing was performed according to the established method by Maiden et al. (3) and isolates were assigned sequence types according to the Neisseria Multi Locus Sequence Typing website (http://pubmlst.org/neisseria/). An additional region in the fumC gene was amplified to determine the presence of a G-to-A point mutation at position 360, characteristic of ET-15 strains (11). All 11 isolates contained this particular point mutation and were therefore classified as ET-15 strains.

PFGE analysis of the 11 isolates that represented the three antigenic variants, C:2a:P1.5,2, C:2a:P1.7,1, and C:2a:P1.5, was performed as described by Tyler and Tsang (9). Restriction enzyme digestion of genomic DNA with NheI (data not shown) and SpeI indicated that serogroup C ET-15 variants had overall similarity and were difficult to distinguish based on the banding patterns they exhibited (Fig. 1). Using SpeI, PFGE pattern I was unique to isolates with the serosubtype P1.5 antigen. However, pattern II was common to all three ET-15 antigenic variants, while pattern III was common to C:2a:P1.5,2 and C:2a:P1.7,1 isolates, and pattern IV was common to C:2a:P1.5,2 and C:2a:P1.5 isolates.

Despite the widespread acceptance of the PFGE method (4, 10), there is an apparent lack of correlation between an isolate's DNA fingerprint and its antigenic profile. This serves to illustrate a potential limitation of PFGE in the analysis of N. meningitidis strains for molecular epidemiology studies of invasive meningococcal disease. Nevertheless, for localized outbreak analysis, PFGE is still a very useful tool to identify strains linked to a common source (5, 6). In summary, a number of typing tools, including both phenotypic and genotypic methods, should be used in combination with carefully documented epidemiological information for surveillance and analysis of meningococcal disease.