Mycoplasma hyopneumoniae Increases Intracellular Calcium Release in Porcine Ciliated Tracheal Cells

Abstract

We investigated the effects of intact pathogenic Mycoplasma hyopneumoniae, nonpathogenic M. hyopneumoniae, and Mycoplasma flocculare on intracellular free Ca²⁺ concentrations ([Ca²⁺]_i) in porcine ciliated tracheal epithelial cells. The ciliated epithelial cells had basal [Ca²⁺]_i of 103 ± 3 nM (n = 217 cells). The [Ca²⁺]_i increased by 250 ± 19 nM (n = 47 cells) from the basal level within 100 s of the addition of pathogenic M. hyopneumoniae strain 91-3 (300 μg/ml), and this increase lasted ∼60 s. In contrast, nonpathogenic M. hyopneumoniae and M. flocculare at concentrations of 300 μg/ml failed to increase [Ca²⁺]_i. In Ca²⁺-free medium, pathogenic M. hyopneumoniae still increased [Ca²⁺]_i in tracheal cells. Pretreatment with thapsigargin (1 μM for 30 min), which depleted the Ca²⁺ store in the endoplasmic reticulum, abolished the effect of M. hyoneumoniae. Pretreatment with pertussis toxin (100 ng/ml for 3 h) or U-73122 (2 μM for 100 s), an inhibitor of phospholipase C, also abolished the effect of M. hyopneumoniae. The administration of mastoparan 7, an activator of pertussis toxin-sensitive proteins G_i and G_o, increased [Ca²⁺]_i in ciliated tracheal cells. These results suggest that pathogenic M. hyopneumoniae activates receptors that are coupled to G_i or G_o, which in turn activates a phospholipase C pathway, thereby releasing Ca²⁺ from the endoplasmic reticulum. Thus, an increase in Ca²⁺ may serve as a signal for the pathogenesis of M. hyopneumoniae.

Mycoplasmas are a large group of diverse prokaryotic species comprising the class Mollicutes. Mycoplasmas lack cell walls, have remarkably small genomes, are phylogenetically related to gram-positive eubacteria, and are the smallest known self-replicating organisms (25, 26, 27). The surfaces of mycoplasmas clearly play a critical role in the interaction of these organisms with their host cells (10, 28, 38). Mycoplasma hyopneumoniae is the etiological agent of mycoplasmal pneumonia in swine, which continues to cause significant economic losses for swine producers. This organism is an extracellular pathogen, and it colonizes in the respiratory epithelium of the pig. The role of M. hyopneumoniae infection in association with other swine respiratory pathogens has gained increased attention (29). For instance, M. hyopneumoniae potentiates porcine reproductive and respiratory syndrome virus-induced pneumonia (33). M. hyopneumoniae induces pneumonia by first damaging the ciliated epithelial cells of the trachea, bronchi, and bronchioles (6, 21, 32). However, the mechanisms underlying M. hyopneumoniae-induced ciliary damage or loss of cilia are not well understood. Recently, a tracheal epithelial cell model that enabled researchers to study the pathogenesis of M. hyopneumoniae strain 91-3 was developed (40).

The adherence of M. hyopneumoniae to ciliated epithelium is necessary to induce colonization of the organism, which results in the loss of cilia (21, 41, 42). Thus, the adherence of mycoplasmas to host cells is an important initial step in the pathogenesis of mycoplasmal diseases. The adherence process is mainly mediated by receptor-ligand interactions (40, 41, 42, 43). Consistent with this finding are the observations that virulent strains of M. hyopneumoniae adhere to cilia of tracheal tissue in vitro, which contrasts with the action of avirulent strains of M. hyopneumoniae (39).

During a study of the mechanisms by which M. hyopneumoniae induces the loss of cilia of respiratory epithelium, it was noted that an increase in the concentration of Ca²⁺ in medium resulted in the loss of cilia (41). Therefore, we hypothesize that M. hyopneumoniae may increase the intracellular free Ca²⁺ concentrations ([Ca²⁺]_i) of ciliated epithelial cells, which serves as an intracellular signal to induce loss of cilia. The present study was undertaken to investigate the M. hyopneumoniae-induced increase of [Ca²⁺]_i in respiratory ciliated epithelial cells and the mechanisms underlying these increases.

MATERIALS AND METHODS

Reagents.

All reagents were obtained from Sigma Chemical (St. Louis, Mo.), except fura-2 acetoxymethyl ester (fura-2AM), which was obtained from Molecular Probes (Eugene, Oreg.), and U-73122, U-73343, and mastoparan 7 (Mas 7), which were obtained from Biomol (Plymouth Meeting, Pa.).

Mycoplasmas.

The following viable intact mycoplasmas were used in the present study: a pathogenic strain, M. hyopneumoniae 91-3, originally cloned from strain 232, which shows high adherence to cilia in a microtiter adherence assay (41); nonpathogenic M. hyopneumoniae strain J (ATCC 25934), which does not adhere to cilia (43); and Mycoplasma flocculare strain Ms42 (ATCC 27399), which is nonpathogenic in swine. Mycoplasmas were cultured in Friis medium (11) to logarithmic phase and harvested by centrifugation at 15,000 × g for 30 min. Following centrifugation, the mycoplasma pellets were collected and washed three times with 50 ml of phosphate-buffered saline (PBS) by centrifugation at 15,000 × g for 15 min. The final pellets were dispersed in PBS through a 27-gauge needle. The number of mycoplasma whole cells collected from 200 ml of culture was determined as color-changing units (CCU) ([3.4 ± 1.7] 10¹¹ × 10¹¹ CCU [mean ± standard deviation]; n = 7) by use of serial dilutions with tubes containing Friis medium. This cell density corresponded to 2.70 ± 0.08 mg of protein measured by the bicinchoninic acid method (Pierce, Rockford, Ill.), as previously described (40, 42). The final mycoplasma concentration in PBS was adjusted to 3 mg of protein/ml.

Tracheal cells.

Cells were isolated as previously described (39). Briefly, the tracheas of 3- to 6-month-old specific-pathogen-free pigs anesthetized with sodium pentobarbital were removed by aseptic techniques. The ciliated cells were dissociated with 0.15% pronase and 0.01% DNase in Ca²⁺- and Mg²⁺-free minimal essential medium, which was incubated at 4°C for 24 h. The epithelial cells were collected by centrifugation at 125 × g for 5 min. The cell pellets were resuspended in a 1:1 mixture of Dulbecco's modified Eagle medium (high glucose) and Ham's F-12 medium containing 5% fetal bovine serum, 0.12 U of insulin per ml, and 100 U of penicillin-streptomycin per ml. Cell suspensions were transferred to 90-mm-diameter tissue culture dishes and incubated in 5% CO₂ for 60 to 90 min to remove fibroblasts. The tracheal epithelial cells were stored in liquid nitrogen until use.

[Ca²⁺]_i measurement in single cells.

The tracheal cells were loaded with 4 μM fura-2AM in Krebs-Ringer bicarbonate buffer solution (136 mM NaCl, 4.8 mM KCl, 1.5 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 10 mM HEPES, 5.5 mM glucose, 0.1% bovine serum albumin [pH 7.4]) and incubated for 30 min at 37°C. The loaded cells were centrifuged at 700 × g for 2 min and then resuspended with Krebs-Ringer bicarbonate at a concentration of 500 to 1,000 cells/ml. The tracheal cells loaded with fura-2AM were plated onto polylysine-coated coverslips in a custom-made petri dish. The dish containing fura-2AM-loaded cells was mounted on the stage of an inverted fluorescence microscope (Carl Zeiss, Thornwood, N.Y.). We focused only on viable ciliated tracheal cells for the determination of [Ca²⁺]_i at 24°C. The fura-2AM-loaded porcine ciliated tracheal cells deteriorated quickly at 37°C. The ciliated cells were selected based on the beating of their active cilia, which indicated that they were alive and viable.

Fluorescence images were obtained (excitation wavelengths of 334 and 380 nm; emission wavelength of 510 ± 20 nm), the background was subtracted, and the images were divided on a pixel-by-pixel basis to generate spatially resolved maps of [Ca²⁺]_i. The emitted signals were digitized, recorded, and processed with an Attofluor digital fluorescence imaging system (Atto Instruments, Rockville, Md.). After fluorescence was read for 150 s, mycoplasmas were mixed with the cell system. The [Ca²⁺]_i were calculated as previously described (13). Calibration was performed in situ according to the procedure provided by Atto Instruments, with fura-2 penta K⁺ as a standard.

Experimental protocols.

To compare how the [Ca²⁺]_i of tracheal cells responded to pathogenic M. hyopneumoniae strain 91-3, avirulent M. hyopneumoniae, and M. flocculare, the cells were treated with these strains at the same concentration (300 μg/ml). One to five ciliated single tracheal cells in each experiment were selected to investigate the changes in [Ca²⁺]_i. These mycoplasmas were maintained on ice before being applied to the tracheal cells.

To investigate the pathway of Ca²⁺ signaling, we preincubated pertussis toxin (PTX) (100 ng/ml) with tracheal cells for 3 h. To deplete the store of Ca²⁺ in the endoplasmic reticulum (ER), the cells were pretreated with 1 μM thapsigargin (TG) for 30 min at 37°C prior to the addition of the mycoplasmas (34). The cells were pretreated with U-73122 (2 μM), a phospholipase C (PLC) inhibitor (2), or its inactive analogue, U-73343, for 100 s at 37°C prior to the addition of the mycoplasmas. To confirm that mycoplasmas increased [Ca²⁺]_i by activating a G_i or G_o protein, Mas 7 (10 μM), an activator of this protein (15), was used to determine if it could increase [Ca²⁺]_i in tracheal cells. In addition, we determined whether PTX could block the increase in [Ca²⁺]_i due to Mas 7. None of the reagents were cytotoxic, since they did not significantly affect the [Ca²⁺]_i of ciliated tracheal cells or decrease or stop the beating of cilia.

Statistical analysis.

Data on [Ca²⁺]_i were analyzed by analysis of variance or Student's t test. The significance level was set at a P value of <0.05.

RESULTS

Effects of mycoplasmas on [Ca²⁺]_i in porcine ciliated tracheal epithelial cells.

Previous studies have demonstrated that M. hyopneumoniae strain 91-3 binds to the cilia of porcine tracheal cells (6, 21, 30). Therefore, the changes in [Ca²⁺]_i were determined after the inoculation of ciliated tracheal cells with strain 91-3. The ciliated epithelial cells had basal [Ca²⁺]_i of 103 ± 3 nM (n = 217 cells). After exposure to M. hyopneumoniae strain 91-3 at a concentration of 300 μg/ml, an increase in [Ca²⁺]_i was observed in 89% (47 of 53 cells in 10 experiments) of the cells. As shown in Fig. 1 and 2, administration of pathogenic M. hyopneumoniae strain 91-3 (300 μg/ml) increased [Ca²⁺]_i in ciliated cells within 100 s. In contrast, nonpathogenic M. hyopneumoniae (inoculated into 18 cells in six experiments) and M. flocculare (inoculated into 24 cells in eight experiments) did not increase [Ca²⁺]_i at the same mycoplasma concentration (300 μg/ml) (Fig. 1).

FIG. 1.

[Ca²⁺]_i responses in ciliated porcine tracheal cells to pathogenic M. hyopneumoniae strain 91-3 (PMH), nonpathogenic M. hyopneumoniae (NPMH), and M. flocculare (MF). Data are means ± standard errors. Intact M. hyopneumoniae 91-3 was administered at concentrations of 30 (n = 18 tracheal cells inoculated in six experiments), 100 (n = 16 cells in seven experiments), and 300 (n = 47 cells in 10 experiments) μg/ml. M. flocculare (n = 24 cells in eight experiments) and nonpathogenic M. hyopneumoniae (n = 18 cells in six experiments) were administered at concentrations of 300 μg/ml. Asterisks indicate significant differences in results for other treatments (P < 0.05).

FIG. 2.

Effects of Ca²⁺-free medium, TG, U-73122, and U-73343 on M. hyopneumoniae-induced increase in [Ca²⁺]_i. Shown are representative graphs of [Ca²⁺]_i in cells after inoculation with M. hyopneumoniae strain 91-3 in the Ca²⁺-free medium (n = 5 cells) (a) and after pretreatment with TG for 30 min (n = 5 cells) (b), 2 μM U-73122 (n = 5 cells) for 100 s (c), or 2 μM U-73343 (n = 5 cells) for 100 s (d). Arrows indicate the administration of intact mycoplasmas (300 μg/ml).

In a dose-response study, a 30-μg/ml concentration of M. hyopneumoniae strain 91-3 (inoculated into 18 cells in six experiments) did not significantly change the [Ca²⁺]_i (Fig. 1). However, M. hyopneumoniae concentrations of 100 μg/ml (inoculated into 16 cells in seven experiments; 84% of the cells responded) and 300 μg/ml (inoculated into 47 cells in 10 experiments; 89% of the cells responded) increased the [Ca²⁺]_i by 110± 9 and 250 ± 19 nM, respectively (Fig. 1).

Effects of M. hyopneumoniae strain 91-3 in Ca²⁺-free medium.

To determine the mechanism by which M. hyopneumoniae strain 91-3 increases [Ca²⁺]_i, we first sought to identify the source of the increase. If extracellular Ca²⁺ plays a role, chelation with EGTA should prevent Ca²⁺ influx. Experiments were performed with Ca²⁺-free medium supplemented with 10 μM EGTA, a Ca²⁺ chelator. M. hyopneumoniae strain 91-3 (300 μg/ml) still increased the [Ca²⁺]_i (from 117 ± 6 to 324 ± 31 nM; 10 cells inoculated in four experiments; 84% of the cells responded) (Fig. 2a). These results suggested that the increase is attributable to Ca²⁺ release from intracellular stores rather than to a Ca²⁺ influx mechanism.

Effect of TG on M. hyopneumoniae-induced [Ca²⁺]_i increase.

To determine whether the ER was the source of Ca²⁺ release, ciliated cells were treated with 1 μM TG, a microsomal Ca²⁺-ATPase inhibitor, for 30 min. In previous studies, TG was found to deplete the ER Ca²⁺ store (33), since it abolished ionomycin-induced intracellular Ca²⁺ release from porcine ciliated tracheal cells (S.-C. Park and W. H. Hsu, unpublished data). Similarly, TG treatment abolished the increase in [Ca²⁺]_i induced by M. hyopneumoniae strain 91-3 (300 μg/ml), suggesting that this organism evokes ER Ca²⁺ release from porcine tracheal epithelial cells (Fig. 2b).

Effects of U-73122 and U-73433 on M. hyopneumoniae-induced [Ca²⁺]_i increase.

Since inositol-1,4,5-trisphosphate (IP₃) releases Ca²⁺ from the ER and IP₃ production is catalyzed by PLC, we determined whether M. hyopneumoniae increases [Ca²⁺]_i through this pathway. Pretreatment of tracheal cells with 2 μM U-73122, a specific PLC inhibitor (2), before inoculation with M. hyopneumoniae strain 91-3, abolished the mycoplasma-induced [Ca²⁺]_i increase in the ciliated cells (Fig. 2c). In contrast, U-73343, an inactive analogue of U-73122, did not prevent the [Ca²⁺]_i response to the mycoplasma (basal concentration of 90 ± 12 nM; peak concentration of 330 ± 25 nM; 10 cells inoculated in four experiments; 82% of the cells responded) (Fig. 2d). These findings suggested that the increase in [Ca²⁺]_i induced by M. hyopneumoniae is mediated by activation of PLC.

Effects of PTX on M. hyopneumoniae- and Mas 7-induced [Ca²⁺]_i increase.

Since intracellular signals are usually mediated by G proteins, we assessed whether a PTX-sensitive G protein mediated the effect of M. hyopneumoniae strain 91-3. In untreated control cells, M. hyopneumoniae strain 91-3 increased [Ca²⁺]_i (254 ± 57 nM; 9 cells in three experiments; 81% of the cells responded) (Fig. 3a). In contrast, pretreatment of ciliated cells with 100 ng of PTX/ml for 3 h abolished M. hyopneumoniae-induced increases in [Ca²⁺]_i (Fig. 3b). These results suggested that M. hyopneumoniae activates receptors that are coupled to a PTX-sensitive G protein (G_i or G_o). To confirm that G_i and G_o proteins are involved in the [Ca²⁺]_i increase in the tracheal cells, we studied the effect of Mas 7, an activator of G_i and G_o (15), on [Ca²⁺]_i. Administration of 10 μM Mas 7 to ciliated tracheal cells evoked an increase in [Ca²⁺]_i from the basal level of 103 ± 4 to 351 ± 24 nM (n = 9 cells in three experiments; 82% of the cells responded) within 100 s (Fig. 3c). Pretreatment of these cells with PTX abolished the effect of Mas 7 (Fig. 3d). These results suggested that activation of G_i or G_o in ciliated tracheal cells increases the [Ca²⁺]_i.

FIG. 3.

Effect of PTX on Mas 7- and intact M. hyopneumoniae-induced increases in [Ca²⁺]_i. Shown are representative graphs of [Ca²⁺]_i in cells after inoculation with M. hyopneumoniae 91-3 and Mas 7, respectively, after pretreatment with PTX (100 ng/ml) for 3 h. (a) M. hyopneumoniae (300μg/ml) controls (n = 5 cells); (b) PTX plus M. hyopneumoniae (n = 5 cells); (c) Mas 7 (10 μM) controls (n= 5 cells); (d) PTX plus Mas 7 (n = 5 cells). Arrows indicate the administration of intact mycoplasmas (300 μg/ml).

DISCUSSION

M. hyopneumoniae colonizes the swine respiratory tract by binding to ciliated epithelial cells (21, 31, 41). Adherence is mediated through the surface protein P97 (15), which has been extensively characterized (16, 17, 22). Ciliostasis and loss of cilia quickly ensue through unknown mechanisms (5). These earlier studies were the first ones to link Ca²⁺ flux with loss of cilia. Pathogenic M. hyopneumoniae strain 91-3 increased the [Ca²⁺]_i in porcine ciliated tracheal cells. Interestingly, nonpathogenic M. hyopneumoniae strain J and M. flocculare failed to do so, indicating that binding to cilia may be a prerequisite for Ca²⁺ flux induction. M. hyopneumoniae strain J does not bind to swine cilia (42). The [Ca²⁺]_i response was a rapid event, and the increase was dependent on the mycoplasma concentration. In study of [Ca²⁺]_i increase induced by M. hyopneumoniae in neutrophils, 10⁷ to 10¹⁰ CCU of the pathogenic strain enhanced the zymosan-induced increase in [Ca²⁺]_i whereas the nonpathogenic strain did not (7). Adherence of pathogenic M. hyopneumoniae strain 91-3 (10⁹ CCU) to the cilia of respiratory epithelia results in the tangling, clumping, and longitudinal splitting of cilia within 90 min of the mycoplasma administration, whereas nonpathogenic M. hyopneumoniae does not cause ciliary damage (6, 39). Therefore, changes in [Ca²⁺]_i in the tracheal epithelia could be a critical mechanism in the pathogenesis of M. hyopneumoniae.

We found that the magnitude of the [Ca²⁺]_i increase in isolated ciliated cells in response to M. hyopneumoniae varied from cell to cell, but in general, [Ca²⁺]_i increased with an increasing concentration of mycoplasmas. This heterogeneity of Ca²⁺ response in the airway epithelial cells was similar to the effect produced by extracellular ATP that was reported for glial cells (36), bile duct cells (23), megakaryocytes (32), and chondrocytes (3). In the respiratory epithelial cells of rabbits, the heterogeneity of the Ca²⁺ response is due to the sensitivity of individual cells to extracellular ATP (9, 20).

High concentrations of mycoplasmas were necessary to increase the [Ca²⁺]_i of ciliated tracheal cells in the present study. Since the tracheal cells used in the present study had not been cultured to ensure recovery, we speculate that the mycoplasma receptors of these cells might have been damaged extensively, thereby requiring high concentrations of mycoplasmas to increase the Ca²⁺ flux. Further studies with cultured tracheal cells will be needed to prove or disprove this hypothesis. Increased [Ca²⁺]_i due to the presence of microorganisms or their toxins have been reported for other bacteria. Intact Salmonella enterica serovar Typhimurium increases [Ca²⁺]_i in intestinal epithelia, which mediates the increase in interleukin-8 secretion from these cells (12, 24). How serovar Typhimurium induces an increase in [Ca²⁺]_i is not yet clear. Escherichia coli enterotoxin elevates [Ca²⁺]_i by releasing ER Ca²⁺ from HEp-2 cells (1). This release is attributable to the activation of ryanodine receptor Ca²⁺ release channels, since the effect is blocked by a ryanodine receptor antagonist, dantrolene (4, 14). Intact verocytotoxin-producing E. coli, however, releases Ca²⁺ from HEp-2 cells via the IP₃ pathway (18). Pyocyanine, a redox-active substance secreted by Pseudomonas aeruginosa, increases IP₃ formation and [Ca²⁺]_i in human airway epithelial cells but reduces G-protein-coupled receptor agonist-induced increases in IP₃ and [Ca²⁺]_i (8). Pyocyanine-induced oxidation may be responsible for the increase in IP₃ formation (8). Pasteurella multocida toxin (PMT) increases [Ca²⁺]_i in different intact animal cells by activating G_q-coupled PLC-β₁ isozyme (37). This effect of PMT is largely attributable to its direct activation of the G_q-PLC pathway, since microinjection of PMT into Xenopus oocytes, which bypasses the plasma membrane receptors, still activates the G_q-PLC pathway (36).

Some extracellular bacterial structures can increase the [Ca²⁺]_i of host cells. For example, type IV pili of pathogenic neisseriae adhere to the epithelial cell-like human cell line ME180 derived from cervical carcinoma and increase [Ca²⁺]_i via the pilus receptors (19). Elevation of [Ca²⁺]_i is needed as an initial step to establish a stable contact between the bacteria and the host cells (19). However, it is not clear how the pili of neisseriae cause an increase in [Ca²⁺]_i. Information on initial [Ca²⁺]_i response by intact bacteria may be important for understanding the pathogenesis.

Pathogenic M. hyopneumoniae strain 91-3 increased [Ca²⁺]_i in the ciliated cells in Ca²⁺-free medium, which suggests that the increase in [Ca²⁺]_i is attributable to Ca²⁺ release from intracellular stores. Pretreatment of tracheal cells with TG to deplete the Ca²⁺ store in the ER abolished the effect of the mycoplasmas, confirming the involvement of this organelle in the Ca²⁺ release. Pretreatment of tracheal cells with U-73122, a specific PLC inhibitor, also prevented the mycoplasma-induced [Ca²⁺]_i increase, suggesting that the mycoplasma-induced Ca²⁺ release from the ER is via a PLC pathway.

Our present findings suggest that the receptors of M. hyopneumoniae in respiratory epithelia are coupled to G_i or G_o, thereby activating PLC, which is consistent with observations with A₁ adenosine receptor-mediated phenomenon (35). To our knowledge, this is the first report showing that activation of G_i or G_o induces an increase in the [Ca²⁺]_i of ciliated tracheal epithelial cells. G_i and G_o proteins are usually responsible for the inhibition of adenylyl cyclase, regulation of K⁺ and Ca²⁺ channels, and activation of cyclic GMP phosphodiesterase. Among G_i and G_o proteins, G_i2 and G_i3 can mediate the modulation of two signaling pathways; activation of PLC is mediated by G_βγ dimer, whereas inhibition of adenylyl cylase is mediated by α_i (35).

In summary, our findings suggest that the receptors for pathogenic M. hyopneumoniae are coupled to G_i or G_o. Once binding of these receptors has occurred, this G protein stimulates the PLC pathway to increase [Ca²⁺]_i through a rise in Ca²⁺ release from the ER. The study of the mechanisms involved in the increase of [Ca²⁺]_i in response to intact mycoplasmas will provide a better understanding of the pathogenesis of mycoplasmal pneumonia. Further studies are warranted to determine which components of the mycoplasma mediate this stimulation in ciliated cells. Along this line, it has been recently discovered that adhesins from M. hyopneumoniae, including P97, failed to increase [Ca²⁺]_i but that an unknown surface protein may mediate this effect of mycoplasma (Park and Hsu, unpublished). This protein and its receptors are currently being characterized. Further studies are also needed to determine the pathophysiological significance of the present findings on the mycoplasma-induced increase in [Ca²⁺]_i. It has been recently discovered that inoculation of porcine ciliated tracheal cells with M. hyopneumoniae strain 91-3 increased the ciliary beating frequency within 3 min of inoculation, which was coupled to the increase in [Ca²⁺]_i in these cells (Park and Hsu, unpublished). These results were consistent with those found with respect to Ca²⁺ action on ciliary beating frequency in ovine airway epithelial cells (30) and suggested the involvement of changes in [Ca²⁺]_i in the pathogenesis of mycoplasmas.