Decontamination of mycoplasma-contaminated Orientia tsutsugamushi strains by repeating passages through cell cultures with antibiotics

Background Mycoplasmas-contamination of Orientia tsutsugamushi, one of the obligated intracellular bacteria, is a very serious problem in in vitro studies using cell cultures because mycoplasmas have significant influence on the results of scientific studies. Only a recommended decontamination method is to passage the contaminated O. tsutsugamushi strains through mice to eliminate only mycoplasmas under influence of their immunity. However, this method sometimes does not work especially for low virulent strains of O. tsutsugamushi which are difficult to propagate in mice. In this study, we tried to eliminate mycoplasmas contaminants from both high virulent and low virulent strains of the contaminated O. tsutsugamushi by repeating passage through cell cultures with antibiotics in vitro. Results We cultured a contaminated, high virulent strain of O. tsutsugamushi using a mouse lung fibroblasts cell line, L-929 cell in the culture medium containing lincomycin at various concentrations and repeated passages about every seven days. At the passage 5 only with 10 μg/ml of lincomycin, we did not detect mycoplasmas by two PCR based methods whereas O. tsutsugamushi continued good growth. During following four passages without lincomycin, mycoplasmas did not recover. These results suggested that mycoplasmas were completely eliminated from the high virulent strain of O. tsutsugamushi. Furthermore, by the same procedures with 10 μg/ml of lincomycin, we also eliminated mycoplasmas from a contaminated, low virulent strain of O. tsutsugamushi. Our additional assay showed that 50 μg/ml of lyncomycin did not inhibit the growth of O. tsutsugamushi, although MICs of many mycoplasmas contaminants were less than 6 μg/ml as shown previously. Conclusion Our results showed an alternative method to eliminate mycoplasmas from the contaminated O. tsutsugamushi strains in place of in vivo passage through mice. Especially this notable method works for the decontamination not only from the high virulent strain also from the low virulent strain of O. tsutsugamushi. For further elimination, lincomycin at the limit concentration, which does not inhibit the growth of O. tsutsugamushi, can possibly eliminate most mycoplasmas from contaminated O. tsutsugamushi strains.


Background
The contamination of cell cultures by mycoplasmas is a serious problem because these bacteria have multiple effects on cell cultures and also have a significant influence on the results of scientific studies. The mycoplasmas are not harmless bystanders and thus cannot be ignored in the cell cultures.
Various elimination methods were previously reported [1][2][3]. These methods are mainly based on four general procedures, physical, chemical, immunological and chemotherapeutic treatment. The physical procedures include heat treatment and filtration. The chemical procedures, treatments to detergents and other chemicals which are effective only against mycoplasmas, but not against host cells. The immunological procedures include in vitro co-culture with macrophages and specific anti-mycoplasmas antisera and in vivo passage thorough mice. The chemotherapeutic procedures are mainly antibiotics treatments that are kills mycoplasmas.
Orientia tsutsugamushi, which is the causative agents of scrub typhus is one of the obligated intracellular bacteria [4]. The mycoplasmas-contaminations of O. tsutsugamushi is also very serious in the in vitro studies using cell cultures. Furthermore the most effective methods for elimination of mycoplasmas can not be applied for decontamination of O. tsutsugamushi strains because these methods also inhibit the growth of O. tsutsugamushi. Decontamination methods should have strong effect on mycoplasmas, but have minimum or no effect on O. tsutsugamushi. Only the recommended decontamination method is to passage the contaminated O. tsutsugamushi strains through mice. Mouse immunity possibly eliminates only mycoplasmas, although O. tsutsugamushi can survive in its target cells, mainly endothelial cells, splenocytes and hepatocytes. In fact, homogenized spleen of infected mice is generally used for the next inoculation. However, this method sometimes does not work especially for low virulent strains of O. tsutsugamushi because they are generally difficult to propagate in mice.
Some of the antibiotics, which are used for elimination of mycoplasmas from tissue culture, are supposed to have less effect against O. tsutsugamushi according to the differences of minimum inhibitory concentrations (MICs) of antibiotics between mycoplasmas [5][6][7] and O. tsutsugamushi [8]. In this study, we tried to eliminate mycoplasmas from contaminated O. tsutsugamushi strains by repeating passages through cell cultures with antibiotics in vitro.

Results and discussion
According to the MICs of antibiotics in the previous reports [5,[7][8][9], we used two antibiotics, lincomycin and ciprofloxacin for elimination of mycoplasmas from the contaminated O.tsutsugamushi strains (Table 1). Both lincomycin and ciprofloxacin are effective against mycoplasmas. Unfortunately there is no available information about the MICs of lincomycin against O. tsutsugamushi. However, according to the MICs of lincomycin against gram-negative bacteria [10], lincomycin is supposed to be much less effective against O. tsutsugamushi because O. tsutsugamushi is one of the gram-negative bacteria. For the example, the MICs of lincomycin against Escherichia coli, one of the typical gram gram-negative bacteria are more than 50 times higher than those against mycoplasmas. Ciprofloxacin was also less effective against O. tsutsugamushi. The MICs of ciprofloxacin against O. tsutsugamushi are about 3 to 200 times higher than those against mycoplasmas (Table 1).
Our result of the direct sequencing showed that Ikeda and Kuroki strains of O. tsutsugamushi were contaminated with Mycoplasma hominis and M. orale respectively. M. hominis and M. orale are 10 to 30% of contaminants of cell cultures ( Table 2) [11]. Previous reports showed that M. fermentas, M. hyorhinis, M. arginini and Acholeplasma laidlawii are the most common contaminants as well as M. hominis and M. orale. More than 90% of the contaminants were caused by these six mycoplasmas [11,12]. The TaqMan PCR and the nested PCR can detect not only all the 6 most common contaminants also some other mycoplasmas. These facts suggested that the detection methods were very reliable to monitor mycoplasmas-contaminations in this study.
For elimination of mycoplasmas, we first cultured a contaminated, high virulent Ikeda strain of O. tsutsugamushi using L-929 cell in the culture medium containing lincomycin and ciprofloxacin and repeated the passages ( Figure 1). Lincomycin and ciprofloxacin were used at 100, 10 and 1 μg/ml. However, ciprofloxacin at 100 μg/ml were cytotoxic against L-929 cell in the first assay and was omitted from the further analyses. We checked mycoplasma-contaminations and O. tsutsugamushi-growth at each passage by the two PCR based methods and/or an immunofluorescent (IF) staining (see Additional file 1). From the passage 1 to 2 with 10 μg/ml of lincomycin, the real-time PCR showed that mycoplasmas decreased, whereas O. tsutsugamushi did not decrease. At the passage 4 with the same concentration of lincomycin, the realtime PCR did not detect mycoplasmas, however the nested PCR still detected them. At the passage 5, both the real-time PCR and the nested PCR did not detect mycoplasmas, whereas the flourish growth of O. tsutsugamushi was observed by IF staining. We continued to culture with lincomycin until the passage 6. During following passages from 7 to 10 without lincomycin, mycoplasmas did not recover. These results clearly showed that mycoplasmas were completely eliminated from O. tsutsugamushiinfected cells. However, the cultivation with 100 μg/ml of lincomycin as well as 10 and 1 μg/ml of ciprofloxacin decreased both mycoplasmas and O. tsutsugamushigrowths, whereas the cultivation with 1 μg/ml of lincomycin did not influence the neither growths.
By the same procedure of Ikeda strain, we cultured a contaminated, low virulent Kuroki strain of O. tsutsugamushi with lincomycin at 10 μg/ml (Figure 1). Mycoplasmas and MICs were obtained from previous reports. a) from [8] and b) from [5][6][7]. c) Gentamycin was not effective against Orinetia tsutsugamushi in a mouse model [25].
O. tsutsugamushi were monitored by the nested PCR and the IF assay respectively (see Additional file 2). At the passage 8, the nested PCR did not detect mycoplasmas. We then continued to cultivate it with lincomycin until the passage 11. During following passages from 12 to 14 without lincomycin, mycoplasmas did not recover. These results showed that we successfully eliminated mycoplasmas also from the low virulent Kuroki strain. The elimination length of Kuroki strain was longer than that of Ikeda strain probably because numbers and/or antibioticssusceptibility of the contaminated mycoplasmas were different. For further elimination of mycoplasmas from other strains of O. tsutsugamushi, we should first evaluate a maximum concentration of lincomycin that does not influence O. tsutsugamushi-growth, and then apply it for decontamination because maximum effects against mycoplasmas are necessary to eliminate them for a short time and to avoid producing lincomycin-resistant mycoplasmas [13][14][15] during repeating passages. Our additional assay showed that lincomycin at 25 μg/ml did not affect the growth (the virulent strain), whereas 50 μg/ml slightly decreased (did not inhibit) the growth in the IF assay (Table 3). Many previous reports about antibioticssusceptibilities of isolated mycoplasmas showed that MICs of lyncomycin against M. hominis, M. fermentas and A. laidlawii, which are the major contaminants, were less  [11,12]. b) This broad-range TaqMan PCR can detect many species of mycoplasmas [22]. c) This nested PCR is highly sensitive, and it is used to check for mycoplasma contamination in the Cell Bank of BioResource Centre, Riken Tsukuba Institute, Tsukuba, Ibaraki, Japan [21]. d) PCR assay for sequencing of mycoplasmas designed in this study. Partial Match means that 2 or 3 of the total of 4 nested-PCR primers match to available regions of the tuf gene on the public database. than 6 μg/ml (0.025 to 6 μg/ml) [5,[16][17][18]. In actual, a previous report showed that lincomycin at 50 μg/ml successfully eliminated the other major contaminants of mycoplasmas, M. hyorhinis and M. hominis from cell cultures [19]. However, a previous report showed that some isolates of M. hyorhinis were highly resistant to lyncomycin (MICs > 100 μg/ml) [14] and a few reports showed that other species of mycoplasmas but not major species of contaminants were highly resistant to lyncomycin [13,15]. Considering these facts, lincomycin at 50 μg/ml can possibly eliminate the contaminants from many of other contaminated strains of O. tsutsugamushi, although it might not be effective for all the cases.

Conclusions
Our results showed an alternative method to eliminate mycoplasmas from the mycoplasma-contaminated strains of O. tsutsugamushi in place of in vivo passage through mice. Especially this new method works for the decontamination not only from the high virulent strain also from the low virulent strain of O. tsutsugamushi, which is difficult to propagate in mice. For further elimination, lincomycin at the limit concentration, which does not inhibit the growth of O. tsutsugamushi, can possibly eliminate most mycoplasmas from contaminated O. tsutsugamushi strains.

Mycoplasmas-contaminated O. tsutsugamushi strains for elimination
A mycoplasmas-contaminated high virulent Ikeda strain and a low virulent Kuroki strain of O. tsutsugamushi were used for elimination. These strains were accidentally contaminated during a long passage history probably because mycoplasmas-contaminated cell culture was used for propagation of these strains. The mycoplasmafree L-929 cell was used for propagation as mentioned in the previous section.

Detection and quantification of mycoplasmas
Major mycoplasmas are listed in Table 2. Upper 6 species are the most common contaminants in cell cultures [11,12]. In order to monitor mycoplasmas, we extracted DNA from O. tsutsugamushi-infected L-929 cell with a commercial DNA extract kit (Tissue genomic DNA extraction mini kit, Favorgen biotech corporation, Ping-Tung, Taiwan) and detected mycoplasmas by two high sensitive and broad range PCR based methods for detection, the nested PCR [21] and the real-time PCR (TaqMan PCR) [22]. The nested PCR is used to check mycoplasma-contaminations in the Cell Bank of Bioresource Centre, Riken Tsukuba institute, Tsukuba, Ibaraki, Japan. For determination of mycoplasma species, we designed new sequencing primers against tuf gene ( Table 2). These designed primers matched tuf gene of 19 mycoplasmas on the public database. All the primers and the probe are listed in Table 4.

Detection of O. tsutsugamushi
To monitor the growth of O. tsutsugamushi, we used a real-time PCR against the gene encoding 47kDa common antigen (Table 4). We extracted DNA from O. tsutsugamushi-infected L-929 cell as mentioned in the previous section and performed the real-time PCR according to the general procedure [23]. We also used an IF staining to monitor the growth of O. tsutsugamushi. In this staining, human convalescent sera of a scrub typhus patient, which were permitted by the ethics committee (number 255), and anti-human antibody conjugated with AlexaFluor W 488 (Life technologies Japan Ltd, Tokyo, Japan) were used. A part of the infected cells were harvested and fixed on a glass slide with ice cold acetone and then the slide was applied for the IF staining according to the previous reports [24].

Antibiotics
Lincomycin (Wako Pure Chemical Industries, Ltd., Osaka, Japan) and ciprofloxacin (Wako Pure Chemical Industries, Ltd., Osaka, Japan) were used for elimination of mycoplasmas in this study. Kanamycin and gentamycin are routinely used for propagation of O. tsutsugamushi to avoid accidental bacterial contamination in our laboratory because they do not influence O. tsutsugamushi-growth [25].

Elimination of mycoplasmas from O. tsutsugamushiinfected cells with antibiotics
We cultured the contaminated strains of O. tsutsugamushi using L-929 cell in the culture medium containing lincomycin and ciprofloxacin at 100, 10 and 1 μg/ml in 25cm2 tissue culture flask, and repeated passages about every seven days. At each passage, the infected cells were harvested. One-third of the harvested cells was used for the next inoculation, another one-third was used for DNA extraction, and the remaining one-third was frozen and stocked. Elimination of mycoplasmas was checked by the nested PCR and/or real-time PCR. The growth of O. tsutsugamushi was monitored by the real-time PCR and/or the IF staining.  Ots-47k-R 5'-AGCAATTCCACATTGTGCTG-3' Ots-47k-P b) 5'-TGCTTAATGAATTAACTCCAGAATT-3' a) Locked nucleic acid (LNA) bases (underlined) and was synthesized with the fluorescent reporter 6-carboxyfluorescein (FAM) covalently coupled to the 5' end and a dark quencher to the 3' end. b) TaqMan probe was synthesized with the fluorescent reporter 6-carboxyfluorescein (FAM) covalently coupled to the 5' end and a dark quencher to the 3' end.