Original Article

Free Access

Mycoplasma bovis isolates from dairy calves in Japan have less susceptibility than a reference strain to all approved macrolides associated with a point mutation (G748A) combined with multiple species-specific nucleotide alterations in 23S rRNA

Toyotaka Sato

Laboratory of Food Microbiology and Food Safety, Department of Health and Environmental Sciences, School of Veterinary Medicine, Rakuno Gakuen University, 582 Bunkyoudai-Midorimachi, Ebetsu, 069-8501, Japan

Department of Microbiology, Sapporo Medical University School of Medicine, S1 W17, Chuo-ku, Sapporo, 060-8556, Japan

Search for more papers by this author

Hidetoshi Higuchi,

Hidetoshi Higuchi

Laboratory of Animal Health, Department of Health and Environmental Sciences, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

Search for more papers by this author

Shin-ichi Yokota,

Shin-ichi Yokota

Department of Microbiology, Sapporo Medical University School of Medicine, S1 W17, Chuo-ku, Sapporo, 060-8556, Japan

Search for more papers by this author

Yutaka Tamura,

Corresponding Author

Yutaka Tamura

[email protected]

Correspondence

Yutaka Tamura, Laboratory of Food Microbiology and Food Safety, Department of Health and Environmental Sciences, School of Veterinary Medicine, Rakuno Gakuen University, 582 Midorimachi Bunkyoudai Ebetsu, Hokkaido 069-8501, Japan. Tel: +81-11-388-4890; fax: +81-11-388-4890; email; [email protected]

Search for more papers by this author

Toyotaka Sato,

Toyotaka Sato

Department of Microbiology, Sapporo Medical University School of Medicine, S1 W17, Chuo-ku, Sapporo, 060-8556, Japan

Search for more papers by this author

Hidetoshi Higuchi,

Hidetoshi Higuchi

Laboratory of Animal Health, Department of Health and Environmental Sciences, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

Search for more papers by this author

Shin-ichi Yokota,

Shin-ichi Yokota

Department of Microbiology, Sapporo Medical University School of Medicine, S1 W17, Chuo-ku, Sapporo, 060-8556, Japan

Search for more papers by this author

Yutaka Tamura,

Corresponding Author

Yutaka Tamura

[email protected]

Correspondence

Search for more papers by this author

First published: 15 May 2017

https://doi.org/10.1111/1348-0421.12490

Citations: 7

Share a link

Email
Wechat
Bluesky

ABSTRACT

Erythromycin, tylosin and tilmicosin are approved for use in cattle in Japan, the latter two being used to treat Mycoplasma bovis infection. In this study, 58 M. bovis isolates obtained from Japanese dairy calves all exhibited reduced susceptibility to these macrolides, this widespread reduced susceptibility being attributable to a few dominant lineages. All 58 isolates contained the G748A variant in both the rrl3 and rrl4 alleles of 23S rRNA, whereas a reference strain (PG45) did not. G748 localizes in the central loop of domain II (from C744 to A753) of 23S rRNA, which participates in binding to mycinose, a sugar residue present in both tylosin and tilmicosin. A number of in vitro-selected mutants derived from M. bovis PG45 showed reduced susceptibility to tylosin and tilmicosin and contained a nucleotide insertion within the central loop of domain II of rrl3 (U747–G748Ins_CU/GU or A743–U744Ins_UA), suggesting that mutations around G748 confer this reduced susceptibility phenotype. However, other Mycoplasma species containing G748A were susceptible to tylosin and tilmicosin. Sequence comparison with Escherichia coli revealed that M. bovis PG45 and isolates harbored five nucleotide alterations (U744C, G745A, U746C, A752C and A753G) in the central loop of domain II of 23S rRNA, whereas other Mycoplasma species lacked at least two of these five nucleotide alterations. It was therefore concluded that G748 mutations in combination with species-specific nucleotide alterations in the central loop of domain II of 23S rRNA are likely sufficient to reduce susceptibility of M. bovis to tylosin and tilmicosin.

List of Abbreviations

MIC: minimum inhibitory concentration
RAPD: random amplified polymorphic DNA
rrnA: A2058G mutation in the rrl3 operon
TMS: tilmicosin

Some species of Mycoplasma, the smallest free-living organisms, are pathogenic in food-producing animals 1. Mycoplasma bovis is a causative agent of bovine respiratory diseases, mastitis and arthritis, causes considerable morbidity and has a substantial negative economic effect on the cattle industry 2. Antimicrobials such as fluoroquinolones, macrolides and tetracyclines are key agents for treating M. bovis infections and for preventing their spread in dairy calves. In Japan, three macrolides (erythromycin, tylosin and tilmicosin) are currently approved for use in cattle. Erythromycin contains a 14-membered lactone ring, whereas tylosin and its derivative tilmicosin contain 16-membered lactone rings 3. Tylosin and tilmicosin are commonly used to treat M. bovis infection in cattle, whereas erythromycin is not. On the basis of its 16S rRNA sequence, M. bovis belongs phylogenetically to the Mycoplasma hominis group 4. M. hominis shares a species-specific genetic background for intrinsic resistance to erythromycin; namely, a G2057A substitution in domain V of 23S rRNA 5. M. bovis PG45, the entire genome of which has been sequenced 6, contains this substitution.

The MICs of tylosin and tilmicosin against M. bovis range from 0.5 µg/mL to >512 µg/mL worldwide 7-14. Having less susceptibility to tylosin and tilmicosin than wild-type M. bovis correlates with the presence of one or two point mutations (G748A, C752T, A2058G, A2059G or A2059C, according to the numbering of the Escherichia coli sequence) in the 23S rRNA alleles rrl3 and rrl4 in M. bovis isolates 15. The A2058G mutation in the rrl3 operon (denoted as rrnA in this article) is associated with reduced susceptibility to both macrolides and lincosamides 16. These findings are consistent with those of genetic studies of susceptibility to tylosin and tilmicosin in other Mycoplasma species, and with the results of basic studies in E. coli 17-19. However, further investigation is required to clarify whether these mutations cause less susceptibility than a reference strain to macrolides in Mycoplasma. Because no ‘gold standard’ direct genetic examination has been available, the biological relevance of genetic mutations that are associated with macrolide resistance in Mycoplasma species is typically determined by in vitro selection of mutants that show little susceptibility, with genomic alignment to corresponding E. coli nucleotide sequences 5, 20-22. Such evaluations have not previously been performed for M. bovis. Investigation of molecular mechanisms underlying less susceptibility than a reference strain to macrolides could provide valuable information to guide strategies for treating M. bovis infections and preventing the spread of resistant M. bovis in cattle.

In this study, the contribution of mutations in the genes encoding 23S rRNA and the ribosomal proteins L4 (RplD) and L22 (RplV) to susceptibility to tylosin and tilmicosin was investigated in M. bovis isolates from dairy calves in Japan. M. bovis mutants with less susceptibility to macrolides than wild-type strains were generated by in vitro selection and characterized. Susceptibility to lincomycin, which has not been approved for use in cattle in Japan, was also investigated. Lincomycin has a similar mode of action to macrolides, but differs structurally. Our results suggest that genetic alterations in M. bovis that combine to confer reduced susceptibility to all three approved macrolides, but not to lincomycin, are widely prevalent.

MATERIALS AND METHODS

Bacterial strains

Analyses were performed on 58 M. bovis isolates that had previously been obtained from nasal swabs of dairy calves from 51 farms (each with ≥100 cattle) located in seven regions of Hokkaido Prefecture in Japan (Dounan, Iburi–Hidaka, Ishikari–Sorachi, Kitami–Okhotsk, Kushiro–Nemuro, Soya–Rumoi–Kamikawa and Tokachi) in 2010–2011 23.

These regions cover ∼83,450 km², and include most of the major dairy-producing areas of Japan. The M. bovis had been isolated and identified previously 23, 24, and then stored until use at −80°C in NK broth (Kanto Chemical, Tokyo, Japan) after cultivation at 37°C in 5% CO₂ for 3 days. M. bovis isolates had been spread on modified Hayflick agar (35.5 g/L pleuropneumonia-like-organism agar base, 15% [v/v] horse serum, 2.5% [w/v] yeast extract, 0.0024% [w/v] deoxyribonucleic acid) (Kanto Chemical) from the stock and cultured at 37°C in a 5% CO₂ atmosphere for 3 days. Colonies were picked up, inoculated into 1 mL NK broth, and cultured at 37°C in a 5% CO₂ atmosphere for 72 hr for antimicrobial susceptibility testing and genetic analysis. Permission had been obtained from all farmers to collect the specimens. The experiments were performed using bacterial strains isolated as part of routine veterinary examination and did not involve any invasive procedures.

Susceptibilities to macrolides and lincomycin

MICs were determined by the agar-dilution method, with modified Hayflick agar and 10⁴–10⁵ CFUs per plate, as described previously 23. The MIC was defined as the lowest concentration of antibiotic that completely inhibited colony formation at 37°C in a 5% CO₂ atmosphere following a 72 hr incubation 23, 25. MIC breakpoints for M. bovis have not yet been defined 26, strains with high MICs (≥64 µg/mL) were therefore defined as having les susceptibility than a laboratory strain, M. bovis PG45 (ATCC25523).

RAPD PCR

Genomic DNA of M. bovis isolates was extracted with a Wizard Genomic DNA Purification Kit (Promega, Fitchburg, WI, USA) from the bacteria in 3 mL of Mycoplasma NK broth after 5 days of culture. RAPD PCR was performed using a KAPA2G Fast HotStart MultiPlex PCR kit (Kapa Biosystems, Wilmington, MA, USA) and the primer Hum4 27. The 25 µL PCR mixture contained 10 ng of template DNA, 1 × KAPA2G Buffer A, 0.2 mM of each dNTP, 0.8 µM Hum4 and 0.5 U of KAPA2G Fast HotStart DNA polymerase. The PCR protocol was as follows: 95°C for 2 min, followed by 40 cycles of denaturation at 95°C for 15 s, annealing at 36°C for 15 s, and extension at 72°C for 20 s and a final extension at 72°C for 7 min. PCR was performed in a GeneAmp PCR System 2400 (PerkinElmer, Waltham, MA, USA), and 5 µL of amplified products was separated by electrophoresis on 1.5% (w/v) agarose gels in 1 × Tris–acetate–EDTA buffer, followed by staining with ethidium bromide for visualization by ultraviolet transillumination. RAPD profiles were digitized for analysis by BioNumerics version 5.10 software (Applied Maths, Sint-Martens-Latem, Belgium). Similarity coefficients were calculated with the Dice algorithm. Clustering was performed by the unweighted pair-group method with arithmetic mean algorithm. The tolerance and optimization levels were 2.0%.

In vitro selection of M. bovis mutants with reduced susceptibility to tylosin and tilmicosin

M. bovis PG45 was cultured in 10 mL of Mycoplasma NK broth for 3 days and then centrifuged at 10,000 rpm for 10 min. Cell pellets were resuspended in 1 mL of 0.85% (w/v) NaCl and then spread (∼10¹⁰ CFU) onto modified Hayflick agar containing 8 µg/mL tylosin or 32 µg/mL tilmicosin (concentrations fourfold higher than the MICs). Cells were cultured at 37°C in a 5% CO₂ atmosphere for 1 week, after which colonies that had grown were subcultured at least twice on agar containing tylosin (8 µg/mL) or tilmicosin (32 µg/mL).

Sequencing of 23S rRNA genes (rrl3 and rrl4) and ribosomal protein genes (rplD and rplV)

Nucleotide sequences of rrl3 and rrl4, rplD, and rplV were determined by PCR and direct DNA sequencing. PCR was performed using AmpliTaq Gold (Thermo Fisher Scientific, Waltham, MA, USA) with the primer pairs listed in Table 1. The PCR primers were designed according to the nucleotide sequence of M. bovis PG45 (GenBank accession number NC_014760). The PCR mixture (50 µL) contained 20 ng of template DNA, 1 × buffer I., 5 mM of MgCl₂, 0.25 mM of each dNTP, 0.8 µM of each primer and 1.25 U of AmpliTaq Gold DNA polymerase. The PCR protocol was as follows: 94°C for 10 min, followed by 40 cycles of denaturation at 94°C for 1 min, annealing at 50°C for 2 min, extension at 72°C for 2 min and a final extension at 72°C for 7 min. PCR was performed in a GeneAmp PCR System 2400. The PCR products were purified by using Wizard SV Gel and PCR Clean-Up system according to the manufacturer's instructions (Promega). Nucleotide sequences of the amplified genes were determined using a BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific) and a 3130 Genetic Analyzer (Thermo Fisher Scientific), with M. bovis PG45 as a reference strain. Accuracy of the prediction of 23S rRNA secondary structure was determined using a bioinformatics tool, CentroidFold 28, 29. Sequences were converted to the E. coli nucleotide numbering by reference to E. coli strain K12 substrain MG1655 (accession number U00096). M. bovis PG45 was the reference for numbering the amino acid substitutions of L4 and L22.

Table 1. Primers used for PCR amplification and sequencing of 23S rRNA genes (rrl3 and rrl4) and ribosomal protein genes (rplD and rplV) in Mycoplasma bovis.

Primer	Sequence (5′–3′)	Position†	Purpose
MBOVPG45_0282-F	ACTTATCTCTAAATTCAAAAGGATATCTAAC	−2209 bp to +399 bp‡	DNA amplification of rrl3
23S-5R	GCCAGGATCAAACTCTCGAAAAA	−2209 bp to +399 bp‡	DNA amplification of rrl3
23S-1F	AGAACGTGGGGATGGATTACCTC	−338 bp to +453 bp§	DNA amplification of rrl4
secY-R	TTAACTGTAGTCATAATAAGGTAGATTGC	−338 bp to +453 bp§	DNA amplification of rrl4
23S-1R	TCTCTACGGCTCCGCTTTTATCTG	647–670 bp¶	Sequencing of rrl3 and rrl4
23S-2F	GAAAGGTGAAAAGAACCCCGAGA	508–530 bp¶	Sequencing of rrl3 and rrl4
23S-2R	CATTGCCCTAAGAACGCTCCTCT	1219–1241 bp¶	Sequencing of rrl3 and rrl4
23S-3F	TCAGCTAAGGTCCCAAAATCGTG	1032–1054 bp¶	Sequencing of rrl3 and rrl4
23S-3R	TCGCATAAGCTAACGCATCCACT	1852–1874 bp¶	Sequencing of rrl3 and rrl4
23S-4F	AAATGACCCCGTAAGTTCGCAAG	1700–1722 bp¶	Sequencing of rrl3 and rrl4
23S-4R	GAACCGACTCCAGCTCCAGGAT	2515–2537 bp¶	Sequencing of rrl3 and rrl4
23S-5F	GGCAGTCGCTCAACGGATAAAAG	2412–2434 bp¶	Sequencing of rrl3 and rrl4
Mb.rplD-F	GCTGAACAAGCTGCACAAGC	−56 bp to +62 bp††	DNA amplification and sequencing of RplD gene encoding L4 protein
Mb.rplD-R	CCTAAGGCATCATTTGACTTTTCA	−56 bp to +62 bp††
Mb.rplV-F	CAACGGTCGTCAACACATCG	−123 bp to +39 bp‡‡	DNA amplification and sequencing of RplV gene encoding L22 protein
Mb.rplV-R	CCATAACGGAAGCCATTTGGA	−123 bp to +39 bp‡‡

†Nucleotide positions of the 23S rRNA operons (rrl3 and rrl4) are based on M. bovis PG45 numbering (accession no. NC_014760.1); ‡nucleotide positions upstream and downstream of rrl3; §nucleotide positions upstream and downstream of rrl4; ¶nucleotide position within rrl3 or rrl4; ††nucleotide positions upstream and downstream of rplD; ‡ ‡nucleotide positions upstream and downstream of rplV.

The 23S rRNA nucleotide-alteration genotypes of M. bovis were compared with those of other Mycoplasma species for which the MICs of tylosin and tilmicosin have been determined. These strains are M. hominis PG21 (accession number NC_013511.1), M. hyopneumoniae J (accession number NC_007295), M. hyorhinis BTS7 (accession number AB182581), M. pneumoniae FH (accession number NC_017504.1) and M. putrefaciens KS1 (accession number NC_015946.1).

RESULTS

Susceptibilities to macrolides and lincomycin and 23S rRNA genotyping of M. bovis isolates from dairy calves in Japan

There was no variation in MICs to each of the macrolides and to lincomycin across the M. bovis isolates (58 strains) derived from dairy calves in Japan (Table 2). The MICs were high (≥64 µg/mL) for tylosin and tilmicosin in these isolates, but low in the wild-type PG45 strain (2 µg/mL and 8 µg/mL, respectively). MICs were high for erythromycin in both PG45 (128 µg/mL) and the isolates (≥128 µg/mL), and low for lincomycin (2–16 µg/mL).

Table 2. 23S rRNA mutation genotypes based on the M. bovis PG 45 nucleotide sequence and macrolide MICs in M. bovis isolates.

23S rRNA mutation genotype	Number of strains	ERY	TYL	TMS	LCM	rrl3	rrl4
		MIC†, μg/mL (MIC range)				23S rRNA mutations‡
Wild-type	PG45§	128	2	8	4	−¶	−
I	48	≥128 (64 to ≥128)	64 (64 to ≥128)	≥128	4 (2–16)	G748A	G748A
II	5	≥128	64	≥128	4 (4–8)	G748A	G748A, C758U
III	2	≥128	≥128	≥128	16	G748A	G748A, C1905A
IV	3	≥128	64 (64 to ≥128)	≥128	8 (8–16)	G748A	G748A, A2107U

†Values are shown as modes; ‡E. coli numbering is used; §M. bovis PG45 was used as the wild-type strain; ¶−, wild-type. ERY, erythromycin; LCM, lincomycin; TMS, tilmicosin; TYL, tylosin. G748 corresponds to G788 in M. bovis numbering, C758 to C798, C1905 to C1904 and A2107 to C2106.

Four genotypes of 23S rRNA mutations (Types I–IV) were identified in the 58 M. bovis isolates (Table 2). All M. bovis isolates contained a common single-nucleotide mutation, G748A, which was present in both 23S rRNA operons, rrl3 and rrl4. Type I isolates (48 strains) did not harbor any 23S rRNA mutations other than G748A. Among Type I isolates, five strains had mutations encoding the amino acid substitution Q93K in L22 (data not shown). Six Type I strains contained multiple amino acid substitutions in L4 (S18T, T43A, A51T, V69A, A70T, E75A, A86T, A101T, V107S, E109_A110InsVKEV) and in L22 (Q93H). These L4 substitutions differ from those identified previously 15. Isolates of Types II–IV contained additional mutations in rrl4: C758U (Type II), C1905A (Type III) and A2107U (Type IV), but no amino acid substitutions in L4 and L22.

On the basis of the results of RAPD PCR with the primer Hum4, PG45 and the isolates were classified into four clusters (A–D in Fig. 1) with a cut-off value of 80% similarity. Approximately 70% (41/58) of the isolates belonged to the major cluster B, which included isolates with 23S rRNA mutation genotypes I, III and IV. Cluster D contained the wild-type PG45, in addition to all five isolates of genotype II.

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

**Genotyping of *M. bovis* isolates by RAPD PCR.**

Genotyping of in vitro-selected PG45-derived mutants with reduced susceptibility to macrolides

Twenty mutants were derived by selection by exposure of PG45 to macrolides; 10 each from supplementation with tylosin and tilmicosin (Table 3). The mutants all had nucleotide mutations in rrl3, enabling classification into four genotypes (and a total of seven subgroups) (Table 3). No mutations were present in rrl4. The mutants formed two groups on the basis of susceptibility to macrolides and lincomycin.

Table 3. 23S rRNA mutation genotypes based on the M. bovis PG45 nucleotide sequence and macrolide MICs in mutants derived from M. bovis PG45 by selection with TYL or TMS.

23S rRNA mutation genotype	TYL	TMS	TYL	TMS	LCM	rrl3	rrl4
	Number of mutants		MIC: μg/mL (MIC range)			23S rRNA mutations†
Wild-type‡	−	−	2	8	4	−§	−
Va	0	1	≥128	≥128	4	U747-G748Ins_CU	−
Vb	0	4	64	≥128	4	U747-G748Ins_GU	−
VI	1	3	≥128	≥128	4	A743-U744Ins_UA	−
VIIa	6	0	≥128	≥128	4	A2062U	−
VIIb	2	0	≥128	≥128	4	A2062C	−
VIIIa	0	2	≥128	≥128	≥64	A2058U	−
VIIIb	1	0	≥128	≥128	≥64	A2058C	−

†E. coli numbering; ‡M. bovis PG45 was used as the wild-type strain; §−, wild-type. LCM, lincomycin; TMS, tilmicosin; TYL, tylosin; A743 corresponds to A783 in M. bovis numbering, U744 to C784, U747 to U787, G748 to G788, A2058 to A2057, and A2062 to A2061.

One group (Types V–VII) contained nine mutants from tylosin-supplemented agar and eight from tilmicosin-supplemented agar; this group had high MICs for tylosin and tilmicosin (≥64 µg/mL), but not for lincomycin (4 µg/mL). Five genotype V (dinucleotide insertion between U747 and G748) mutants were only obtained from tilmicosin-supplemented agar, four genotype VI (dinucleotide insertion between A743 and U744) mutants were obtained from tylosin and tilmicosin supplementation, and eight genotype VII (single-nucleotide substitution at A2062) mutants were only obtained from tylosin-supplemented agar.

The other group contained three Type VIII mutants with high MICs for tylosin and tilmicosin (≥128 µg/mL) and for lincomycin (≥64 µg/mL). The genotype VIII (single-nucleotide substitution at A2058) mutants were obtained from tylosin and tilmicosin supplementation.

DISCUSSION

All M. bovis isolates (58 strains) examined in this study exhibited less susceptibility than a reference strain to all macrolides (erythromycin, tylosin and tilmicosin) approved for treatment of cattle in Japan; however, they were susceptible to lincomycin, a non-approved lincosamide (Table 2). Similar macrolide-susceptibility patterns have been described in previous studies on large and diverse sets of M. bovis isolates in Japan 7, 10, indicating that the currently approved macrolides are likely to be unsuccessful against M. bovis infection in dairy cattle in Japan. Moreover, we have previously demonstrated that 12% (7/58) of these M. bovis isolates are also resistant to fluoroquinolones 23. These results have implications for future selection of antimicrobials for use in the Japanese dairy industry, which are likely to become progressively more restricted.

The high prevalence of M. bovis isolates exhibiting low susceptibility to tylosin and/or tilmicosin has also been reported in some other countries, including the USA 13, UK 8, Hungary 11 and France 9, whereas ∼50% of isolates are susceptible to tylosin and tilmicosin in Israel, including isolates from domestic cattle and from cattle imported from Europe and Australia 15. Notably, susceptibility of M. bovis to most antibiotics seems to have decreased markedly over the past few decades 9. These findings indicate that M. bovis strains with little susceptibility to macrolides have spread worldwide. We suggest that a small number of dominant lineages of M. bovis with less susceptibility than a reference strain to macrolides have spread throughout the Japanese dairy cattle and now have an extremely high prevalence.

Available evidence suggests that low susceptibility of M. bovis isolates to macrolides with 16-membered-ring structures (including tylosin and tilmicosin) is associated with a G748A mutation in both 23S rRNA operons (rrl3 and rrl4) 15. It also suggests that combinations of other mutations (C752T, A2058G and A2059G/C) with G748A are associated with higher MIC values for both tylosin and tilmicosin than individual mutations 15. In the present study, we detected G748A in both rrl3 and rrl4 in all isolates, suggesting that this mutation has an important role in reducing the susceptibility of M. bovis to tylosin and tilmicosin . By contrast, we do not consider that the observed amino acid substitutions in L4 and L22 or some mutations in rrl4 (encoding C758U, C1905A and A2107U alterations in the RNA) contribute to the low susceptibility to tylosin and tilmicosin because we did not observe these changes in all isolates with high MICs.

Some strains of other Mycoplasma species, such as M. hyopneumoniae J, M. hyorhinis BTS7 and M. hominis PG21, harbor the G748A mutation but remain susceptible to 16-membered-ring macrolides, including tylosin and tilmicosin (Table 4, Fig. 2). In addition, a point mutation at G748 only increases the tylosin MIC 2-fold in E. coli; additional mutations being required for the acquisition of reduced susceptibility to tylosin and tilmicosin 30. These data suggest that the G748A substitution is necessary but not sufficient for reducing susceptibility to tylosin and tilmicosin, and that additional factors contribute to this phenotype in M. bovis isolates.

Table 4. Nucleotide alterations in the central loop of domain II of 23S rRNA genes and susceptibility to TYL and TMS in various Mycoplasma species.

Nucleotide position†	M. bovis PG45	M. bovis isolates from Japan	M. hyopneumoniae J M. hyorhinis BTS7 M. hominis PG21	M. pneumoniae FH	M. putrefaciens KS1
	Mycoplasma strain
U744	U744C	U744C	U744C	U744C	U744C
G745	G745A	G745A	–	–	–
U746	U746C	U746C	U746C	–	–
G748	–	G748A	G748A	–	–
A752	A752C	A752C	–	A752C	–
A753	A753G	A753G	A753G	A753G	A753G
TYL	S	R	S	S	S
TMS	S	R	S	n.d.	n.d.

†Nucleotide sequence was compared with that of the E. coli reference strain using the following accession numbers: M. hyopneumoniae J (NC_007295), M. hyorhinis BTS7 (AB182581), M. hominis PG21 (NC_013511.1), M. pneumoniae FH (NC_017504.1) and M. putrefaciens KS1 (NC_015946.1). n.d., not determined; R, reduced susceptibility (MIC ≥64 µg/mL); S, susceptible (MIC ≤32 µg/mL); TMS, tilmicosin; TYL, tylosin. Data for susceptibility were taken from previous studies: M. hyopneumoniea J 39; M. hyorhinis BTS7 21; M. hominis PG21 5; M pneumoniae FH 5 and M. putrefaciens KS1 40.

To detect relevant factors other than G748A, we investigated nucleotide alterations in domain II of 23S rRNA of M. bovis (from C744 to G753). This region participates in binding to mycinose, a sugar residue that is present in 16-membered-ring macrolides, such as tylosin and tilmicosin, but absent in other macrolides and lincosamides 3, 31. Nucleotide alterations in this region that suppress interaction with 16-membered-ring macrolides should lead to reductions in susceptibility to them. In the present study, we compared the nucleotide sequences of M. bovis, other Mycoplasma species and E. coli with the aim of identifying any species-specific nucleotide alterations in M. bovis (Table 4). We found that the reference strain PG45 and the M. bovis isolates shared a combination of five nucleotide alterations (U744C, G745A, U746C, A752C and A753G) in the central loop of domain II in both alleles of 23S rRNA that are not present in E. coli (Table 4, Fig. 2). This combination of nucleotide alterations has not previously been reported. We found that other Mycoplasma species lack at least two of the alterations and are susceptible to tylosin and tilmicosin (Table 4, Fig. 2), suggesting that the nucleotide alterations in the central loop of domain II of 23S rRNA are associated with reduced susceptibility to tylosin and tilmicosin. In a previous study, the point mutation G745A occurred during in vitro exposure of M. hyorhinis BTS7 to tylosin, but did not alter the tylosin MIC 21. Although mutation or methylation at G745, G748 or A752 increases tylosin MIC in E. coli, each individual alteration only results in a twofold increase in tylosin MIC 30-32. These lines of evidence suggest that the M. bovis isolates in the present study have evolved reduced susceptibility to tylosin and tilmicosin through occurrence of a common point mutation (G748A) in a background of species-specific nucleotide alterations from the conserved sequence (G745A and A752C, and also perhaps U744C, U746C and A753G) in the central loop of domain II of 23S rRNA. Although we did not fully exclude other factors, such as activation of efflux pumps (which is known to contribute to resistance to other antimicrobials in other Mycoplasma species) 33, our results led us to propose possible molecular mechanisms for the reduced susceptibility of M. bovis to 16-membered-ring macrolides.

Generation and characterization of 20 M. bovis mutants derived from the reference strain PG45 by in vitro selection provided further information on the determinants of antimicrobial susceptibility (Table 3). We found that some of these mutants carry chromosomal alterations that have resulted in nucleotide insertions in the central loop of domain II in rrl3, but not in rrl4, and predicted that these mutations cause drastic changes in the secondary structure of this loop (Fig. 2). These mutations, which contributed to reduced susceptibilities to tylosin and tilmicosin, have not previously been reported in Mycoplasma species. These results provide further evidence that mutations in the central loop of domain II (from A743 to G748) in at least one 23S rRNA allele can lead to reduction in tylosin and tilmicosin susceptibilities.

We found that other mutants generated by in vitro selection harbor mutations at A2062 in rrl3 and exhibit high MICs for tylosin and tilmicosin, but not for lincomycin (Table 3). These mutations have not previously been detected in M. bovis isolates; however, mutations at A2062 contribute to resistance to 16-membered-ring macrolides in M. hominis without affecting susceptibilities to lincomycin and other macrolides 20. These mutations represent an alternative mechanism for reducing susceptibility to 16-membered-ring macrolides; however, that is not currently being observed in M. bovis isolates from dairy cattle in Japan.

We found that a final group of mutants generated by in vitro selection harbor rrl3 mutations at A2058U/C and are associated with high MICs for all the macrolides and lincomycin. A2058, the major binding site for ketolides, macrolides (including those with ring structures containing 14, 15 and 16 members) and lincosamides, forms part of the central loop of domain V of bacterial 23S rRNA 17, 34, 35. Point mutations at A2058 alter the susceptibility of several Mycoplasma species to these agents 22, 36, 37 and A2058G mutations in one or both 23S rRNA alleles have been detected in M. bovis isolates 15, 16. We did not detect A2058G mutations in M. bovis isolates from dairy cattle in Japan in the present study, suggesting that these isolates are still likely to be susceptible to 15-membered-lactone-ring macrolides, ketolides and lincosamides.

Combinations of mutations and nucleotide alterations in 23S rRNA are considered to be associated with reduced susceptibility to macrolides. Point mutations at A2058 in domain V of 23S rRNA of Campylobacter jejuni are associated with low susceptibility to all macrolides, ketolides and lincosamides, but at the expense of a reduction in bacterial fitness 38. G748A mutations in domain II of 23S rRNA may lead to resistance to tylosin and tilmicosin with less effect on fitness than mutations in domain V. M. bovis contains five intrinsic nucleotide alterations (G745A, A752C, U744C, U746C and A753G) in the central loop of domain II of 23S rRNA that are likely to influence binding of tylosin and tilmicosin. Additionally, the G748A mutation could interact with these existing alterations such as to confer resistance to macrolides. The G748A mutation, species-specific nucleotide alterations in the central loop of domain II and the inherited erythromycin-resistant mutation G2057A in domain V of 23S rRNA 5 confer resistance to all macrolides currently approved for use in Japan. The widespread prevalence of low susceptibility to this class of antimicrobial agents has important implications for the treatment of infections in the Japanese dairy industry. Because antimicrobial susceptibility testing is time consuming and difficult for making routine diagnoses in veterinary fields, it may be preferable to institute detection of G748A for rapid estimation of susceptibility of M. bovis isolates to tylosin and tilmicosin. Further study is need to clarify this.

ACKNOWLEDGEMENTS

This study was supported by a grant from a program for developing support systems for upgrading education and research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology and a grant from the Japanese Ministry of Agriculture, Forestry and Fisheries.

DISCLOSURE

The authors declare that this research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.

REFERENCES

1 Rosengarten R., Citti C., Glew M., Lischewski A., Droesse M., Much P., Winner F., Brank M., Spergser J. (2000) Host-pathogen interactions in mycoplasma pathogenesis: Virulence and survival strategies of minimalist prokaryotes. Int J Med Microbiol 290: 15–25.
10.1016/S1438-4221(00)80099-5
CAS PubMed Web of Science® Google Scholar
2 Gagea M.I., Bateman K.G., Van Dreumel T., Mcewen B.J., Carman S., Archambault M., Shanahan R.A., Caswell J.L. (2006) Diseases and pathogens associated with mortality in Ontario beef feedlots. J Vet Diagn Invest 18: 18–28.
10.1177/104063870601800104
PubMed Web of Science® Google Scholar
3 Debono M., Willard K.E., Kirst H.A., Wind J.A., Crouse G.D., Tao E.V., Vicenzi J.T., Counter F.T., Ott J.L., Ose E.E., Omura S. (1989) Synthesis and antimicrobial evaluation of 20-deoxo-20-(3,5-dimethylpiperidin-1-yl)desmycosin (tilmicosin, EL-870) and related cyclic amino derivatives. J Antibiot 42: 1253–67.
10.7164/antibiotics.42.1253
CAS PubMed Web of Science® Google Scholar
4 Pettersson B.U.M., Johansson K.E. (1996) Phylogeny of some mycoplasmas from ruminants based on 16S rRNA sequences and definition of a new cluster within the hominis group. Int J Syst Bacteriol 46: 1093–8.
10.1099/00207713-46-4-1093
CAS PubMed Web of Science® Google Scholar
5 Pereyre S., Gonzalez P., De Barbeyrac B., Darnige A., Renaudin H., Charron A., Raherison S., Bebear C., Bebear C.M. (2002) Mutations in 23S rRNA account for intrinsic resistance to macrolides in Mycoplasma hominis and Mycoplasma fermentans and for acquired resistance to macrolides in M. hominis. Antimicrob Agents Chemother 46: 3142–50.
10.1128/AAC.46.10.3142-3150.2002
CAS PubMed Web of Science® Google Scholar
6 Wise K.S., Calcutt M.J., Foecking M.F., Roske K., Madupu R., Methe B.A. (2011) Complete genome sequence of Mycoplasma bovis type strain PG45 (ATCC 25523). Infect Immun. 79: 982–3.
10.1128/IAI.00726-10
CAS PubMed Web of Science® Google Scholar
7 Uemura R., Sueyoshi M., Nagatomo H. (2010) Antimicrobial susceptibilities of four species of Mycoplasma isolated in 2008 and 2009 from cattle in Japan. J Vet Med Sci 72: 1661–3.
10.1292/jvms.10-0165
CAS PubMed Web of Science® Google Scholar
8 Ayling R.D., Rosales R.S., Barden G., Gosney F.L. (2014) Changes in antimicrobial susceptibility of Mycoplasma bovis isolates from Great Britain. The Vet Rec 175: 486.
10.1136/vr.102303
CAS PubMed Web of Science® Google Scholar
9 Gautier-Bouchardon A.V., Ferre S., Le Grand D., Paoli A., Gay E., Poumarat F. (2014) Overall decrease in the susceptibility of Mycoplasma bovis to antimicrobials over the past 30 years in France. PloS ONE 9: e87672.
10.1371/journal.pone.0087672
PubMed Web of Science® Google Scholar
10 Kawai K., Higuchi H., Iwano H., Iwakuma A., Onda K., Sato R., Hayashi T., Nagahata H., Oshida T. (2014) Antimicrobial susceptibilities of Mycoplasma isolated from bovine mastitis in Japan. Anim Sci J 85: 96–9.
10.1111/asj.12144
CAS PubMed Web of Science® Google Scholar
11 Sulyok K.M., Kreizinger Z., Fekete L., Hrivnak V., Magyar T., Janosi S., Schweitzer N., Turcsanyi I., Makrai L., Erdelyi K., Gyuranecz M. (2014) Antibiotic susceptibility profiles of Mycoplasma bovis strains isolated from cattle in Hungary, Central Europe. BMC Vet Res 10: 256.
10.1186/s12917-014-0256-x
PubMed Web of Science® Google Scholar
12 Lysnyansky I., Ayling R.D. (2016) Mycoplasma bovis: Mechanisms of resistance and trends in antimicrobial susceptibility. Front Microbiol 7: 595.
10.3389/fmicb.2016.00595
PubMed Web of Science® Google Scholar
13 Rosenbusch R.F., Kinyon J.M., Apley M., Funk N.D., Smith S., Hoffman L.J. (2005) In vitro antimicrobial inhibition profiles of Mycoplasma bovis isolates recovered from various regions of the United States from 2002 to 2003. J Vet Diagn Invest 17: 436–41.
10.1177/104063870501700505
PubMed Web of Science® Google Scholar
14 Gerchman I., Levisohn S., Mikula I., Lysnyansky I. (2009) In vitro antimicrobial susceptibility of Mycoplasma bovis isolated in Israel from local and imported cattle. Vet Microbiol 137: 268–75.
10.1016/j.vetmic.2009.01.028
CAS PubMed Web of Science® Google Scholar
15 Lerner U., Amram E., Ayling R.D., Mikula I., Gerchman I., Harrus S., Teff D., Yogev D., Lysnyansky I. (2014) Acquired resistance to the 16-membered macrolides tylosin and tilmicosin by Mycoplasma bovis. Vet Microbiol 168: 365–71.
10.1016/j.vetmic.2013.11.033
CAS PubMed Web of Science® Google Scholar
16 Kong L.C., Gao D., Jia B.Y., Wang Z., Gao Y.H., Pei Z.H., Liu S.M., Xin J.Q., Ma H.X. (2016) Antimicrobial susceptibility and molecular characterization of macrolide resistance of Mycoplasma bovis isolates from multiple provinces in China. J Vet Med Sci 78: 293–6.
10.1292/jvms.15-0304
CAS PubMed Web of Science® Google Scholar
17 Douthwaite S., Champney W.S. (2001) Structures of ketolides and macrolides determine their mode of interaction with the ribosomal target site. J Antimicrob Chemother 48 Suppl T1: 1–8.
10.1093/jac/48.suppl_2.1
PubMed Web of Science® Google Scholar
18 Douthwaite S., Hansen L.H., Mauvais P. (2000) Macrolide-ketolide inhibition of MLS-resistant ribosomes is improved by alternative drug interaction with domain II of 23S rRNA. Mol Microbiol 36: 183–93.
10.1046/j.1365-2958.2000.01841.x
CAS PubMed Web of Science® Google Scholar
19 Zaman S., Fitzpatrick M., Lindahl L., Zengel J. (2007) Novel mutations in ribosomal proteins L4 and L22 that confer erythromycin resistance in Escherichia coli. Mol Microbiol 66: 1039–50.
10.1111/j.1365-2958.2007.05975.x
CAS PubMed Web of Science® Google Scholar
20 Furneri P.M., Rappazzo G., Musumarra M.P., Di Pietro P., Catania L.S., Roccasalva L.S. (2001) Two new point mutations at A2062 associated with resistance to 16-membered macrolide antibiotics in mutant strains of Mycoplasma hominis. Antimicrob Agents Chemother 45: 2958–60.
10.1128/AAC.45.10.2958-2960.2001
CAS PubMed Web of Science® Google Scholar
21 Kobayashi H., Nakajima H., Shimizu Y., Eguchi M., Hata E., Yamamoto K. (2005) Macrolides and lincomycin susceptibility of Mycoplasma hyorhinis and variable mutation of domain II and V in 23S ribosomal RNA. J Vet Med Sci 67: 795–800.
10.1292/jvms.67.795
CAS PubMed Web of Science® Google Scholar
22 Wu C.M., Wu H., Ning Y., Wang J., Du X., Shen J. (2005) Induction of macrolide resistance in Mycoplasma gallisepticum in vitro and its resistance-related mutations within domain V of 23S rRNA. FEMS Microbiol Lett 247: 199–205.
10.1016/j.femsle.2005.05.012
CAS PubMed Web of Science® Google Scholar
23 Sato T., Okubo T., Usui M., Higuchi H., Tamura Y. (2013) Amino acid substitutions in GyrA and ParC are associated with fluoroquinolone resistance in Mycoplasma bovis isolates from Japanese dairy calves. J Vet Med Sci 75: 1063–5.
10.1292/jvms.12-0508
CAS PubMed Web of Science® Google Scholar
24 Higuchi H., Iwano H., Gondaira S., Kawai K., Nagahata H. (2011) Prevalence of mycoplasma species in bulk tank milk in Japan. Vet Rec 169: 442.
10.1136/vr.d5331
CAS PubMed Web of Science® Google Scholar
25 Hannan P.C. (2000) Guidelines and recommendations for antimicrobial minimum inhibitory concentration (MIC) testing against veterinary mycoplasma species. International research programme on comparative mycoplasmology. Vet Res 31: 373–95.
10.1051/vetres:2000100
CAS PubMed Web of Science® Google Scholar
26Clinical and Laboratory Standards Institute. (2015) Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals; Approved Standard-Fourth Edition and Supplement. VET01A4E and VET01S3E. W. Pennsylvania, USA:CLSI.
Google Scholar
27 Mcauliffe L., Kokotovic B., Ayling R.D., Nicholas R.A. (2004) Molecular epidemiological analysis of Mycoplasma bovis isolates from the United Kingdom shows two genetically distinct clusters. J Clin Microbiol 42: 4556–65.
10.1128/JCM.42.10.4556-4565.2004
CAS PubMed Web of Science® Google Scholar
28CentroidFold. [Cited….] Available from URL: http://www.ncrna.org/centroidfold
Google Scholar
29 Hamada M., Sato K., Kiryu H., Mituyama T., Asai K. (2009) Predictions of RNA secondary structure by combining homologous sequence information. Bioinformatics 25: i330–8.
10.1093/bioinformatics/btp228
CAS PubMed Web of Science® Google Scholar
30 Novotny G.W., Jakobsen L., Andersen N.M., Poehlsgaard J., Douthwaite S. (2004) Ketolide antimicrobial activity persists after disruption of interactions with domain II of 23S rRNA. Antimicrob Agents Chemother 48: 3677–83.
10.1128/AAC.48.10.3677-3683.2004
CAS PubMed Web of Science® Google Scholar
31 Liu M., Douthwaite S. (2002) Resistance to the macrolide antibiotic tylosin is conferred by single methylations at 23S rRNA nucleotides G748 and A2058 acting in synergy. Proc Natl Acad Sci 99: 14658–63.
10.1073/pnas.232580599
CAS PubMed Web of Science® Google Scholar
32 Liu M., Kirpekar F., Van Wezel G.P., Douthwaite S. (2000) The tylosin resistance gene tlrB of Streptomyces fradiae encodes a methyltransferase that targets G748 in 23S rRNA. Mol Microbiol 37: 811–20.
10.1046/j.1365-2958.2000.02046.x
CAS PubMed Web of Science® Google Scholar
33 Raherison S., Gonzalez P., Renaudin H., Charron A., Bébéar C., Bébéar C.M. (2002) Evidence of active efflux in resistance to ciprofloxacin and to ethidium bromide by Mycoplasma hominis. Antimicrob Agents Chemother 46: 672–9.
10.1128/AAC.46.3.672-679.2002
CAS PubMed Web of Science® Google Scholar
34 Hansen L.H., Mauvais P., Douthwaite S. (1999) The macrolide-ketolide antibiotic binding site is formed by structures in domains II and V of 23S ribosomal RNA. Mol Microbiol 31: 623–31.
10.1046/j.1365-2958.1999.01202.x
CAS PubMed Web of Science® Google Scholar
35 Roberts M.C., Sutcliffe J., Courvalin P., Jensen L.B., Rood J., Seppala H. (1999) Nomenclature for macrolide and macrolide-lincosamide-streptogramin B resistance determinants. Antimicrob Agents Chemother 43: 2823–30.
10.1128/AAC.43.12.2823
CAS PubMed Web of Science® Google Scholar
36 Matsuoka M., Narita M., Okazaki N., Ohya H., Yamazaki T., Ouchi K., Suzuki I., Andoh T., Kenri T., Sasaki Y., Horino A., Shintani M., Arakawa Y., Sasaki T. (2004) Characterization and molecular analysis of macrolide-resistant Mycoplasma pneumoniae clinical isolates obtained in Japan. Antimicrob Agents Chemother 48: 4624–30.
10.1128/AAC.48.12.4624-4630.2004
CAS PubMed Web of Science® Google Scholar
37 Gerchman I., Levisohn S., Mikula I., Manso-Silvan L., Lysnyansky I. (2011) Characterization of in vivo-acquired resistance to macrolides of Mycoplasma gallisepticum strains isolated from poultry. Vet Res 42: 90.
10.1186/1297-9716-42-90
CAS PubMed Web of Science® Google Scholar
38 Luangtongkum T., Shen Z., Seng V.W., Sahin O., Jeon B., Liu P., Zhang Q. (2012) Impaired fitness and transmission of macrolide-resistant Campylobacter jejuni in its natural host. Antimicrob Agents Chemother 56: 1300–8.
10.1128/AAC.05516-11
CAS PubMed Web of Science® Google Scholar
39 Vicca J., Stakenborg T., Maes D., Butaye P., Peeters J., De Kruif A., Haesebrouck F. (2004) In vitro susceptibilities of Mycoplasma hyopneumoniae field isolates. Antimicrob Agents Chemother 48: 4470–2.
10.1128/AAC.48.11.4470-4472.2004
CAS PubMed Web of Science® Google Scholar
40 Antunes N.T., Tavio M.M., Mercier P., Ayling R.D., Al-Momani W., Assuncao P., Rosales R.S., Poveda J.B. (2007) In vitro susceptibilities of Mycoplasma putrefaciens field isolates. Antimicrob Agents Chemother 51: 3452–4.
10.1128/AAC.00420-07
CAS PubMed Web of Science® Google Scholar

Citing Literature

Volume61, Issue6

June 2017

Pages 215-224

Mycoplasma bovis isolates from dairy calves in Japan have less susceptibility than a reference strain to all approved macrolides associated with a point mutation (G748A) combined with multiple species-specific nucleotide alterations in 23S rRNA