In vitro protective efficacy of Lithium chloride against Mycoplasma hyopneumoniae infection.

Mycoplasma hyopneumoniae (M. hyopneumoniae) infection affects the swine industry. Lithium chloride (LiCl), is a drug used to treat bipolar disorder and has also shown activity against bacterial and viral infections. Herein, we evaluated the antibacterial activity of LiCl on PK-15 cells infected with M. hyopneumoniae. Incubation of LiCl (40mM) with cells for 24h, did not significantly affect the cell viability. The qRT-PCR showed ~80% reduction in M. hyopneumoniae genome when LiCl added post-infection. A direct effect of LiCl on bacteria was also observed. However, treatment of cells with LiCl prior infection, does not protect against the infection. Anti-bacterial activity of LiCl was further confirmed by IFA, which demonstrated a reduction in the bacterial protein. With 40mM LiCI, the apoptotic cell death, production of nitric oxide and superoxide anion induced by M. hyopneumoniae, were prevented by ~80%, 60% and 58% respectively. Moreover, caspase-3 activity was also reduced (82%) in cells treated with 40mM LiCl. LiCl showed activity against various strains of M. hyopneumoniae examined in our study. Collectively, our data showed that LiCl inhibited the infection of M. hyopneumoniae through anti-apoptotic mechanism.

Mycoplasma hyopneumoniae (M. hyopneumoniae) causes porcine enzootic pneumonia (PEP) that affects the swine industry (Clark et al., 1991). Treatment of PEP is limited to antibiotics and vaccines that provide partial protection (Haesebrouck et al., 2004). Development of therapeutic agents against the infection is required.

Lithium chloride (LiCl) is a drug used to treat bipolar disorder (Bowden et al., 2005). It modulates several biological functions, including apoptosis, glycogen synthesis and inflammation (Machado-Vieira et al., 2009). LiCl active against viruses including avian coronavirus (Li et al., 2009) and transmissible gastroenteritis virus (TGEV) (Ren et al., 2011) in addition to bacteria such as Francisella tularensis (Zhang et al., 2009) and Pseudomonas aeruginosa (Chen et al., 2013). However, it un-clear whether LiCl could also inhibit the infection of M. hyopneumoniae.

Reactive oxygen species (ROS) (Beere and Green, 2001), nitric oxide (NO) (Brüne, 2003) and caspases, are important apoptosis inducers. LiCl was found to prevent apoptosis by inhibiting production of NO (Wang et al., 2013), ROS (Lumetti and Galli, 2014) and caspases (Ren et al., 2011). M. hyopneumoniae infection was reported to induce apoptosis (Bai et al., 2013). Thus, we aimed to investigate the inhibitory properties of LiCl on PK-15 cells infected with M. hyopneumoniae and to underline the possible mechanism.

Swine tracheal epithelial cells (STEC) are the ideal model to study Mycoplasma in vitro. However, porcine kidney-15 (PK-15) cells model used in our study was previously established in our laboratory with time and titer above 4 h and 1 × 105  CCU/ml (Che et al., 2013, Che et al., 2014, Che et al., 2012, Zhang et al., 2013) and was also described by others (Burnett et al., 2006, Wang et al., 2016).

M. hyopneumoniae isolates Xiaoling Wei-1(XLW-1), Nanjing (NJ) and attenuated strain (168 L) were grown in KM2 medium (Bai et al., 2013). PK-15 cells were cultured in DMEM containing 10% FBS.

Cytotoxicity of LiCl (MW = 42.39, Sigma (U.S.A.) on PK-15 cells was analyzed by MTT. We observed that LiCl at concentrations and time used in our had no toxic effect on PK-15 cells (data not shown).

Cells (1 × 105 cells/well) in 24-well plates were infected with M. hyopneumoniae (1 × 106  CFU/ml) at 37 °C for 10 h, washed and inoculated with LiCl (10–40 mM) for 12 h. After which, the cells were washed, total DNA was extracted and the effect of LiCl was evaluated by qRT–PCR following previous methods established in our laboratory (Wu et al., 2012). Briefly, the genome sequence of M. hyopneumoniae (GenBank CP002274.1) was used to PCR amplify 460 bp of p97 gene (GenBank: ADQ90328.1). This fragment was cloned in pMD18-T vector (pMD18-T-P97) to generate a log–linear standard curve, from which the DNA copy numbers was determined by qRT–PCR (based on Ct value). The qRT–PCR was performed on ABI PRISM 7500 Sequence Detection System (Applied Biosystems, USA) with SYBR® Premix Ex TaqTM Kit (Takara). The qRT–PCR mixture (25 μl in total) consisted of 12.5 μl Premix Ex TaqTM (2 ×), 0.5 μl ROX Reference Dye (50 ×), 1 μl Forward primer (10 μM), 1 μl Reverse primer (10 μM), 0.125 μl Probe, 9 μl ddH2O and 1 μl DNA while the cycling condition was set at 95 °C for 10 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 s. An about 90 bp of p97 gene was detected using p97 primers Forward- 5′ CCAGAACCAAATTCCTTCGCTG 3′and Reverse- 5′ ACTGGCTGAACTTCATCTGGGCTA 3′and p97 probe 5′ FAM-AGCAGATCTTAGTCAAAGTGCCCGTG-TAMRA 3′ which labeled at the 5' with the reporter dye 6-carboxyfluorescein (FAM) and at the 3' with a Black-Hole-Quencher.

We found that LiCl induce a dose-dependent decrease in the amount of bacterial DNA. At 40 mM, LiCl caused approximately 80% reduction in the relative amplification of p97 gene of M. hyopneumoniae (Fig. 1A). This was consistent with the inhibitory effect of LiCl on DNA synthesis of herpes simplex virus (Ziaie and Kefalides, 1989). Previous studies shown that M. hyopneumoniae introduced into PK-15 has infected the cell after 10 h (Che et al., 2014). Herein, we added LiCl to cells 10 hrs post-infection, indicating that, the inhibition of LiCl occurred during bacterial replication. We further investigated the direct effect of LiCl on cells or bacteria. LiCl-treated cells (1 × 105 cells/well) were infected with untreated bacteria (1 × 106  CFU/ml), and untreated cells were infected with LiCl-treated bacteria (1 × 106  CFU/ml). A direct effect of LiCl on bacteria was also observed but not on cells (Fig. 1B and Fig. C) which is consistent with the effect of LiCl on IBV (Harrison et al., 2007).

Fig. 1

Antibacterial activity of LiCl against M. hyopneumoniae (1 × 106 CCU/ml) infected PK-cells (1 × 105 cells/well). Cells infected with M. hyopneumoniae, XLW-1 for 10 h and treated with LiCl for 12 h. (A) DNA extracted and quantified by qRT–PCR (log10 copies of DNA/well). Data presented as Means ± SD (n = 3). ⁎p < 0.008 and ⁎⁎p < 0.001 vs. control. (B) M. hyopneumoniae treated with LiCl (direct inactivation) before inoculating into cells. (C) Cells treated with LiCl then infected with M. hyopneumoniae, XLW-1 (pre-infection). DNA was extracted and quantified by qRT–PCR as above. A direct effect on bacteria was observed but not in cells. Data presented as Means ± SD (n = 3) ⁎p < 0.05 and p < 0.54 vs. control. (D) Effect of LiCl treatment was evaluated by IFA. Cells washed, fixed and probed with primary (p46 monoclonal antibody) and secondary-FITC conjugated antibodies. FITC-fluorescence was measured at OD 535. ⁎⁎P < 0.001 vs. control.

We further infected cells (1 × 105 cells/well) in 96-well plates with M. hyopneumoniae (1 × 106  CFU/ml) at 37 °C for 10 h. Cells washed and inoculated LiCl (10–40 mM) for 12 h. Expression level of M. hyopneumoniae surface antigen (p46) was evaluated by IFA using p46 monoclonal antibody (our laboratory) and FITC-conjugated goat anti-mouse secondary antibodies (Boster Biological Technology Company, China). Fluorescence was measured immediately at excitation 485 nm and emission of 535 nm using an infinite F200 Tecan Microplate Reader (Huppert et al., 2010), and the data were acquired using i-control software (Tecan Trading AG, Ma¨nnedorf, Switzerland). We also observed that LiCl inhibited p46 expression (Fig. 1D) which indicated that LiCl treatment led to substantial inhibition of M. hyopneumoniae infection.

It has been reported that, M. hyopneumoniae infection could induce apoptosis in PBMC and STEC cells (Bai et al., 2015, Ni et al., 2014). As the infection model of M. hyopneumoniae with PK-15 cells was already developed in our laboratory, it's un-clear whether PK-15 cells undergoes apoptosis upon infection with this bacterium. In this study, by using AO/EB staining (Normal/Apoptotic/Necrotic Cell Detection kit, Beyotime Biotech., China), we observed that, the infection of PK-15 cells with this bacterium could also induce apoptosis in PK-15 cells (Fig. 2A). We further investigated the ability of LiCl to prevent M. hyopneumoniae induced apoptosis. Briefly, cells (1 × 105 cells/well) on 96-wells plate infected with M. hyopneumoniae (1 × 106  CFU/ml) for 10 h, washed and treated with or without LiCl (10–40 mM) for 12 h. Cells washed again, added 35 μl of AO/EB solution, examined by fluorescence microscopy and the apoptosis was quantified. Number of positive cells were calculated by selection of five view fields. Apoptotic rate was calculated as follows: apoptotic rate (%) = number of positive staining cells/number of total cells × 100%. Viable cells appeared green with intact nuclei. Nonviable cells had bright orange chromatin. Apoptotic cells appeared shrinked with condensation and fragmentation of nuclei. Apoptotic cells were distinguished from necrotic cells because the latter appeared orange with a normal nucleus (Renvoize et al., 1998). As shown in (Fig. 2A), M. hyopneumoniae induced apoptosis were significantly decreased in a dose-dependent manner after treatment with LiCl. Counting apoptotic cells at 24 hpi, confirmed this (Fig. 2B) and indicated ~ 80% inhibition of apoptosis at 40 mM LiCL. These results imply that LiCl maybe a universal anti-apoptotic agent with limited selectivity to cell types and apoptosis inducers. Although the result is contradictory with studies on Burkholderia pseudomallei (Maniam et al., 2013) in which pretreatment of macrophages and A549 epithelial lung cells inhibited the infection, these data on post-infection treatment indicate that whether LiCl executes an anti-apoptotic or a pro-apoptotic effect is largely associated to the cell type and stimulating factor.

Fig. 2

Identification of anti-apoptotic mechanism of LiCl against PK-15 cells (1 × 105 cells/well) infected with M. hyopneumoniae (1 × 106 CCU/ml). PK-15 cells were infected with M. hyopneumoniae, XLW-1 for 10 h and treated with or without LiCl for 12 h. (A) Phenotype of apoptosis and anti-apoptotic effect of LiCl were analyzed by AO/EB staining. + Ve denotes for untreated cells while − Ve denotes for neither infected nor treated cells. Most representative fields are shown. (B) Calculated apoptosis rate is provided. Data presented as Means ± SD (n = 3).⁎⁎P < 0.001 vs. control. (C) LiCl treatment inhibits production of NO and (D) superoxide anion induced by M. hyopneumoniae infection. NO production and superoxide anion were evaluated by Griess assay and superoxide anion assay respectively in PK-15 cells infected with M. hyopneumoniae and treated with LiCl. Data presented as mean ± SD (n = 3). ⁎P < 0.05 and ⁎⁎⁎P < 0.001 vs. control. (E) LiCl inhibited caspase-3 activity. Caspase-3 activity was measured in PK-15 cell lysate infected with ~ 1 × 106 CCU/ml M. hyopneumoniae for 10 h prior to LiCl treatment. LiCl inhibited caspase-3 activity in a dose-dependent manner. Data presented as Means ± SD (n = 3). ⁎⁎P < 0.01 vs. control.

Low concentrations of NO inhibits apoptosis, and excessive production of NO triggers apoptosis (Shen et al., 1998). We evaluated whether LiCl inhibits production of NO in PK-15 cells infected with M. hyopneumoniae by using NO assay kit (Beyotime Biotech., China). Cells (1 × 105 cells/well) infected with M. hyopneumoniae (1 × 106  CFU/ml) and treated with or without LiCl (10–40 mM). Supernatants were mixed with Griess-I and Griess-II and examined at 570 nm. NO concentrations in the supernatants was reduced by 60% at 40 mM LiCl (Fig. 2C) and this might correlate with the cells protection.

Superoxide anion is involved in the early apoptosis. We treated cells (1 × 105 cells/well) with or without LiCl post-infection and the level of ROS was measured by a superoxide assay kit (Beyotime Biotech., China). Compared to controls, addition of LiCl reduced ROS production by 58% (Fig. 2D).This inhibitory activity of LiCl against NO and ROS was previously observed (Lumetti and Galli, 2014, Wang et al., 2013). Inhibitors of GSK3β increase CREB (cAMP Response Element-Binding Protein) DNA binding activity and IL-10 expression, that reduce the production of inducible nitric oxide synthase (iNOS) and NO cells (Bécherel et al., 1995, Seyler et al., 1997). Different results were reported regarding LiCl treatment. LiCl was found to potentiate LPS-induced iNOS expression in astrocytes, but not in mouse macrophages (Feinstein, 1998), and LiCl alone slightly increased iNOS expression in hepatocytes and in vivo (Du et al., 2006). These variable findings may signify stimulus or cell type-selective regulatory effects of GSK3β inhibitors on iNOS expression. However, majority of reports likely indicate attenuation of NO production by GSK inhibitor. Excessive ROS lead to mitochondrial dysfunction and activates caspases and apoptosis (Lu et al., 2011), suggesting that LiCl might act through the mitochondrial pathway.

M. hyopneumoniae infection could activate caspase-3 (Bai et al., 2015, Bai et al., 2013). By using caspase-3 activity assay kit (Beyotime Biotech., China), we evaluated whether LiCl blocks M. hyopneumoniae activated caspase-3. Cells (1 × 105 cells/well) infected for 10 h as indicated above and treated with or without LiCl (10–40 mM) for 12 h. Cells lysed in lysis buffer (~ 108  cells/ml), centrifuged and the supernatant was quantified by Bradford method (Beyotime Biotech, China). An about 50 μl of cells lysate (100 μg) was mixed with 40 μl reaction buffer and 10 μl caspase-3 substrate (Ac-DEVD-pNA). After incubating at 37 °C for 2 h, caspase-3 activity was measured at 405 nm. As shown in (Fig. 2E), M. hyopneumoniae infection significantly increased caspase-3 activity. When PK-15 cells treated with LiCl, caspase-3 activity was significantly reduced by 82%, suggesting that LiCl may act upstream of caspase-3 activation. LiCl induced cells neuro-protection against glutamate was found to be associated with up-regulation of anti-apoptotic protein Bcl-2, down-regulation of pro-apoptotic proteins p53 and Bax, and suppressed release of cytochrome c from mitochondria (Ghribi et al., 2002), thereby preventing caspase-3 activation and apoptosis (Adams and Cory, 1998). Thus, anti-apoptotic mechanism of LiCl may also be available in other cell protection mechanisms, including the prevention of M. hyopneumoniae-induced apoptosis.

Various M. hyopneumoniae strains (186L, NJ and XLW-1) were tested for their susceptibility to LiCl. Cells (1 × 105 cells/well) were infected with these strains (1 × 106  CFU/ml) and treated with LiCl. At 40 mM LiCl, the apoptosis was significantly inhibited by 80% (data not shown) in cells infected with these strains which indicate that, the inhibition by LiCl may be a general feature of M. hyopneumoniae or other Mollicutes.

In conclusion, our data indicated that LiCl inhibited M. hyopnemoniae infection on PK-15 cells via antiapoptotic mechanism. These results are potentially relevant to the search for antimicrobials against M. hyopneumoniae infections, as well as other Mollicutes.

Competing interests

The authors declare that they have no competing interests.