Molecular Typing of Mycoplasma pneumoniae Strains in Sweden from 1996 to 2017 and the Emergence of a New P1 Cytadhesin Gene, Variant 2e

Abstract

Mycoplasma pneumoniae causes respiratory infections, such as community-acquired pneumonia (CAP), with epidemics recurring every 3 to 7 years. In 2010 and 2011, many countries experienced an extraordinary epidemic peak. The cause of these recurring epidemics is not understood, but decreasing herd immunity and shifts in the strains' antigenic properties have been suggested as contributing factors. M. pneumoniae PCR-positive samples were collected between 1996 and 2017 from four neighboring counties inhabited by 12% of Sweden's population. A total of 578 isolates were characterized directly from 624 clinical samples using P1 typing by sequencing and multilocus variable number tandem repeat analysis (MLVA). A fluorescence resonance energy transfer (FRET)-PCR approach was also used to detect mutations associated with macrolide resistance in the 23S rRNA gene. Through P1 typing, the strains were classified into type 1 and type 2, as well as variants 2a, 2b, 2c, and a new variant found in nine of the strains, denoted variant 2e. Twelve MLVA types were distinguished, and 3-5-6-2 (42.4%), 4-5-7-2 (37.4%), and 3-6-6-2 (14.9%) predominated. Several P1 and MLVA types cocirculated each year, but type 2/variant 2 strains and MLVA types 3-5-6-2 and 4-5-7-2 predominated during the epidemic period comprising the peak of 2010 and 2011. In 2016 and 2017, type 1 became more common, and MLVA type 4-5-7-2 predominated. We also found that 0.2% (1/578) of the strains carried a macrolide resistance-associated mutation, indicating a very low prevalence of macrolide resistance in this region of Sweden.

Keywords: MLVA; Mycoplasma pneumoniae; P1 typing; molecular typing.

FIG 1

Proportion of PCR-positive M. pneumoniae samples per year in Sweden from 1996 to 2018. The black line is based on results from Gävle County (corresponding to approximately 1,000 to 1,500 samples analyzed per year). The dotted line is based on results from 13 counties in Sweden from 2011 to 2016 (corresponding to approximately 21,000 to 32,000 analyzed samples per year) and four counties in Sweden from 2017 and 2018 (corresponding to approximately 7,000 analyzed samples per year).

FIG 2

Distributions of P1 and MLVA types during different epidemic periods. (a) Distribution between type 1 and type 2/variant 2 strains; (b) distribution between type 2 and different variant 2 strains; (c) distribution of MLVA types.

FIG 3

Differences of the amino acid sequences, which are part of the repMP2/3 within the P1 protein, between type 1 (M18639), variant 1 (AF290000.1), type 2 (AF290001.1), variant 2a (AB024618.1), variant 2b (DQ383277.1), variant 2c (JN048895), variant 2d (EF656612), and the newly discovered variant 2e (MK330954). The amino acid positions correspond to the P1 gene of M. pneumoniae, M129 (NC_000912.1).

FIG 4

Minimum spanning tree of 578 M. pneumoniae isolates from four counties in Sweden from 1996 to 2017. Each circle denotes an MLVA type. The size of the circle is proportional to the number of isolates belonging to each MLVA type, and the color corresponds to the P1 type. The distance between MLVA types corresponds to the number of allelic changes, and each line represents one allelic change.

FIG 5

Schematic figure of the major differences between the repMP2/3 elements situated within the P1 adhesion gene. Variants 2a, 2b, and 2e are compared to type 2 according to the position of strain Mac (CP010550.1). The gray bars indicate the inserted parts of the repMP2/3-a element that characterize each of the variant strains. The insertions are identical to the corresponding positions: 128382 to 128906 (variant 2b, DQ383277.1), 128382 to 128954 (variant 2e, MK330954), and 128688 to 128954 (variant 2a, AB024618.1), except for 12 bp, which shows a deletion at positions 128751 to 128762 (indicated by a black mark) in variants 2a and 2e.