Mycoplasma hyorhinis (M. hyorhinis) is a commensal pathogen of swine that also causes lung lesions and inflammation (Razin et al. 1998), and is thought to contribute to the development of cell transformation in vitro (Namiki et al. 2009). These properties of M. hyorhinis have increased interest to the researchers.

Whole genome sequence of M. hyorhinis strain HUB-1 was determined (Liu et al. 2010), and expression of foreign antigens in M. hyorhinis might help to produce recombinant engineered strains. However, a method based on GFP expressing plasmids to evaluate the transformation and expression of foreign genes in M. hyorhinis has not been described. Several methods to monitor gene activity in cells are available such as the formation of fusion proteins with coding sequences for β-galactosidase, firefly luciferase, and bacterial luciferase (Stewart and Williams 1992). But, these methods are of limited use since they require exogenous substrates or cofactors. The green florescence proteins (GFP) of jellyfish Aequorea victoria is a unique tool to monitor gene transfer and expression (Cubitt et al. 1995). Using GFP might help to construct an efficient reporter system for M. hyorhinis. Here, we constructed a plasmid expressing GFP fluorescence and optimized conditions for transformation by electroporation.

M. hyorhinis strain HUB-1 (GenBank accession CP002170.1) was provided by Prof. Xiao Shaobo (Huazhong Agricultural University, China) and was grown at 37 °C in KM2 cell-free liquid medium (a modified Friis medium) containing 20 % (v/v) swine serum (Xiong et al. 2016). KM2-Agar was prepared by adding 0.7 % Agar (Biowest Agarose ®G-10; Gene Company Limited, Chi Wan, Hong Kong) to KM2 medium and was incubated at 37 °C to grow the visible colonies. Tetracycline hydrochloride (Sigma-Aldrich) was used at 0.01 μg/ml.

We previously constructed a plasmid pMD18-TOgfp encoding tetracycline resistance gene (tetM) controlled by the p97 gene promoter, GFP gene also controlled by the p97 gene promoter and oriC of M. hyopneumoniae attenuated strain (168L) (GenBank accession 507382422) (Ishag et al. 2016). The purpose of this plasmid was to express GFP in M. hyopneumoniae strain 168L. It is well known that, the p97 gene functions as an adhesion molecule for M. hyopneumoniae and the activity of this promoter was previously described in oriC-plasmids of M. hyopneumoniae (Maglennon et al. 2013). Here, we further evaluated the potential of this promoter in M. hyorhinis.

The presence of the oriC in plasmids is necessary to maintain the plasmid in the host, and for mycoplasmas, the oriC has been found to be species specific (Cordova et al. 2002). To construct a specific system expressing GFP in M. hyorhinis, we predicted the oriC of M. hyorhinis strain HUB-1 (Fig. 1a) following previously methods described in M. hyoneumoniae (Maglennon et al. 2013). The oriC was PCR amplified from the DNA of M. hyorhinis (Fig. 1b) using oriC primers listed in Table 1 and was used to replace the oriC of M. hyoneumoniae in the vector pMD18-TOgfp at EcoRI and XhoI restriction sites. The resulting plasmid specific for M. hyorhinis was designated pMD18-O/MHRgfp. The diagram of the initial cloning and introduction of a new oriC is shown in (Fig. 1c). The cloning was verified by restriction enzyme digestion and DNA sequence analysis.

Fig. 1
figure 1

Construction of pMD18-O/MHRgfp. a The predicted origin of replication (oriC) of M. hyorhinis (strain HUB-1). b PCR amplified oriC fragment (1935 bp). c Diagram and map of pMD18-O/MHRgfp construction. The oriC of M. hyopneumoniae strain 186L was removed from pMD18-TOgfp with EcoRI and XhoI restriction enzymes and replaced with oriC of M. hyorhinis. Numbers 1, 2, 3 and 4 indicated the order of the fragments cloned into the vector

Table 1 Primers used to amplify the oriC of M. hyorhinis to construct pMD18-O/MHRgfp plasmid

Transformation of M. hyorhinis by polyethylene glycol (PEG) was reported (Dybvig and Alderete 1988). Here, we optimized methods for transformation by electroporation (Maglennon et al. 2013): We obtained no clones in the KM2-Agar plate containing 0.01 µg/ml tetracycline hydrochloride when we used low voltage (1–1.5 kV) or low concentrations of plasmid DNA (1–5 µg). However, increasing the voltage directly to 2.5 kV and the amount of plasmid DNA to 15 µg could produce 4 × 10−3 cfu/µg plasmid DNA. Briefly, 40 ml of M. hyorhinis culture were centrifuged at 12,000 rpm for 20 min at 4 °C, and the pellet was washed three times with electroporation buffer (272 mM Sucrose, 200 mM HEPES pH 7.2) supplemented with 1 mM EDTA. The product was incubated on ice for 5 min, and resuspended in 100 µl of electroporation buffer. Plasmid DNA (15 µg) was added to 100 µl competent cells and transferred to chilled 0.2 cm electroporation cuvette (Bio-Rad, USA). The mixture was incubated on ice for 20 min. The cells were electroporated on ECM® 630 Electroporation System, BTX™ at 2.5 kV, 125 Ω and 25 µF. Immediately after electroporation, 900 µl of chilled KM2 medium was added and incubated for 20 min on ice and then recovered for 3 h at 37 °C. The culture was diluted, plated on KM2 plates containing 0.7 % Agar and 0.01 µg/ml of tetracycline hydrochloride and incubated at 37 °C until growth of visible clones. Tetracycline-resistant colonies of transformed mycoplasmas grown on KM2-Agar had appeared within 3–10 days of incubation (Fig. 2a). These colonies were absent in the control mycoplasmas that were not electroporated with plasmid.

Fig. 2
figure 2

Detection of pMD18-O/MHRgfp in the transformed cells. M. hyorhinis were transformed with 15 µg pMD18-O/MHRgfp and grown in KM2 medium containing 0.01 µg/ml tetracycline hydrochloride. The same transformants were also plated in KM2-Agar containing 0.01 µg/ml tetracycline hydrochloride and the plate was incubated for 7 days. a Colonies of M. hyorhinis observed under microscopy (scale bar 100 µM). b Transformed mycoplasmas either subcultured in KM2 medium containing 0.01 µg/ml tetracycline hydrochloride or plated in KM2-Agar containing 0.01 µg/ml tetracycline hydrochloride and a single clone picked and sub-cultured in KM2 medium containing 0.01 µg/ml tetracycline hydrochloride. DNA was extracted from control un-transformed mycoplasmas or from transformed mycoplasmas and presence of pMD18-O/MHRgfp was evaluated by PCR to detect GFP from KM2 (b.A) or from KM2-Agar (b.B). The GFP product of about 750 bp was only detected from the DNA of the transformants mycoplasmas and it was consistent with the predicted sizes of GFP. This band was absent from the controls (un-transformed mycoplasmas)

Tetracycline-resistant mycoplasmas were analyzed for their plasmid content. Total genomic DNA was extracted using a TIANamp Bacteria DNA Kit (Tiangen, Beijing, China) from either the pool of mycoplasma cultures containing 0.01 µg/ml tetracycline hydrochloride or from a single resistant clone sub-cultured in KM2 medium containing 0.01 µg/ml tetracycline hydrochloride. The presence of pMD18-O/MHRgfp was analyzed by the detection of GFP (750 bp) by PCR, and GFP could be detected from the total genomic DNA of the transformants, but not from untransformed mycoplasmas (Fig. 2b). One product amplified with GFP specific primers was sequenced and was indeed the expected GFP sequence (data not shown).

Expression of GFP in a single clone of M. hyorhinis selected on KM2-Agar was also studied. Seven day-old colonies showed green fluorescence when observed by fluorescence microscopy (Nikon, Eclipse E600, Tokyo, Japan) (Fig. 3) and this color was absent in the controls. The expression of GFP in M. hyorhinis cells did not appear to interfere with cell growth. Therefore, GFP should also be a vital marker of transformation and cell growth as the pure cultures bearing genetic markers can ease the direct identification of cells and colonies among the population of culture. In related studies, the GFP gene was described as an efficient marker for studying the development and microbe-plant interaction in the tobacco pathogen Phytophthora parasitica var. nicotianae (Bottin et al. 1999). We hypothesize that, tagging M. hyorhinis with a plasmid expressing GFP may help to follow the infection process by in vivo imaging if M. hyorhinis stably harbored the transformed constructs.

Fig. 3
figure 3

Detection of GFP fluorescence. M. hyorhinis was transformed with pMD18-O/MHRgfp, plated in KM2-Agar medium containing 0.01 µg/ml tetracycline hydrochloride and cultured for 7 days. Control mycoplasmas were transformed with a pMD18-T empty plasmid. a Expression of GFP gene upon transformation with pMD18-O/MHRgfp plasmid was observed under the fluorescence microscope. b The green fluorescence protein was not observed in the control mycoplasmas transformed with pMD18-T empty plasmid. Scale bar 100 µM

In the present report, the construction of a vector carrying the GFP gene was performed in order to develop a direct method for monitoring gene transfer and expression in M. hyorhinis in which the timing, as well the magnitude of gene expression, is being examined. This visual expression analysis system could also indicate that, the expression of the heterologous genes in M. hyorhinis is feasible.