Mastitis causes the greatest economic loss to the dairy industry. It has been estimated that costs associated with mastitis for the US dairy industry exceed $2 billion per year [1,2]. Antibiotic treatments and other mastitis management strategies have improved the control of contagious pathogens; nonetheless, mastitis caused by microorganisms is still a major problem, even in well-managed dairy farms [3,4].

Mycoplasmas are highly contagious organisms and can cause various diseases in humans and animals [5]. In dairy cows, Mycoplasma bovis (M. bovis) is known to be the most common and virulent bovine Mycoplasma species in the United States [6].The financial losses of M. bovis infections, as a result of loss of weight gain and carcass value and other associated diseases, are estimated to be approximately $108 million per year with infection rates of up to 70% in a herd [7]. Mycoplasma bovis is characteristically refractory to Beta lactam antibiotics because it does not possess a cell wall. Furthermore, evidence is accumulating that antibiotics, including tetracycline, tilmicosin and spectinomycin, are not effective for the treatment of M. bovis [8]. Thus, finding alternative yet effective treatments against M. bovis is vital.

Extensive research has shown that the various fatty acids have antimicrobial effects and may be used as an inhibitory agent against bacteria [9-11]. Kelsey et al. [12] demonstrated that linoleic acid inhibited growth of two different mastitis strains of Staphylococcus aureus (S. aureus). Prostaglandin F_2α (PGF_2α) is synthesized via a metabolic pathway commencing with arachidonic acid which is previously derived from linoleic acid [13]. Results from a recent study in our laboratory [14,15] indicate that the fatty acid PGF_2α, in the form of dinoprost tromethamine, inhibits the growth of Streptococcus uberis and S. aureus, in a dose dependent manner. These findings resemble the actions of fatty acids in a previously described study [12]. Mycoplasma bovis is phylogenetically related to gram positive bacteria [16] and therefore, PGF_2α may have similar inhibitory effects on growth as previously observed with S. aureus. To our knowledge there is no information on the effect of any prostaglandin on Mycoplasma species. It was hypothesized that PGF_2α has inhibitory effects on the growth of M. bovis associated with bovine mastitis. Thus, the objective was to determine the effect of PGF_2α on M. bovis growth in vitro.

Bacteria strains, experimental design, and growth culture

Five strains of M. bovis from bovine origin were selected for the study. Two strains were reference strains (ATCC 25025 and 25523). The other three strains were isolated from diseased cattle, two from milk of cows with clinical mastitis (MKB and CS-UI) and one from a swabbing solution of a cow’s nasal passage with clinical pneumonia (VP-UI). Mycoplasma bovis isolates were purified, cultured and suspended in sterile phosphate buffered saline to achieve an optical density of 0.2 at 520 nm, yielding approximately 1 × 10⁷ to 2 × 10⁸ CFU/ml as previously described [17]. Subsequently, M. bovis suspensions were incubated in culture media containing PGF_2α (dinoprost tromethamine) at final concentrations of 0 (control), 2, 4, and 8 mg/ml, for 8 h at 37°C in duplicate, where two sample tubes per treatment concentration per strain of Mycoplasma species were tested. After the 8 h incubation a sample from each tube was obtained, serially diluted, and cultured on three modified Hayflick’s agar plates for 10 d and bacterial growth assessed as CFU/ml. The CFU/ml counts from each of the three agar plates were averaged for each corresponding tube. The entire experiment was repeated on 4 different days to account for variation associated with a day effect, categorized as run.

Statistical analysis

Data were analyzed by least squares analysis of variance using GLM procedures of SAS and the model included the effect of experimental run (4 runs), treatment (4 doses of PGF_2α) , strain (5 strains), and their interaction.

Based on CFU/ml, PGF_2α, in the form of dinoprost tromethamine, had inhibitory effects on M. bovis growth; mean CFU/ml decreased (P<0.01) with increasing concentrations of PGF_2α and 8 mg/ml of PGF_2α being the most inhibitory (Figure 1). Mean CFU/ml (106/ml) was significantly decreased from 43.6 ± 1.1 in control (0 mg/ml PGF_2α) to 24.3 and 7.8 ± 1.1, in 4 and 8 mg/ml PGF_2α treatment (Figure 1).

Figure 1: Effect of various doses of prostaglandin F2a on Mycoplasma bovis growth across all five strains1. *There was an effect of treatment dose on mean CFU/ml of M. bovis . Across all strains, mean CFU/ml of M. bovis was less (P < 0.05) for 4 and 8 mg/ml prostaglandin F2a compared to 0 mg/ml. The error bars indicate the standard errors associated with the mean. 1ATCC 25025 and 25523 were reference strains, MKB=isolated from milk of cows with clinical mastitis, CS-UI=isolated from milk of cows with clinical mastitis, VP-UI=swabbing solution of a cow’s nasal passage with clinical pneumonia.

However, an effect of treatment by strain interaction on mean CFU/ml was detected (P<0.05) providing evidence that M. bovis growth across all PGF_2α doses were not similar among strains. Pre-planned contrasts were conducted to compare the mean log CFU/ml values between strains for each treatment. Overall, the 2 mg/ml PGF_2α decreased (P<0.05) CFU/ml only in one strain compared with control, whereas 4 and 8 mg/ml PGF_2α decreased (P<0.01) CFU/ml in all strains compared with control (Figure 2).

Figure 2: Effect of various doses of prostaglandin F2a on each strain of Mycoplasma bovis growth1. There was an effect (P< 0.05) of treatment by strain interaction on mean CFU/ml. 1ATCC 25025 and 25523 were reference strain, MKB=isolated from milk of cows with clinical mastitis, CS-UI=isolated from milk of cows with clinical mastitis, VP-UI=swabbing solution of a cow’s nasal passage with clinical pneumonia. *Different from 0 mg/ml treatment within strain (P< 0.05). The error bars indicate the standard errors associated with the mean.

Our results provide evidence, for the first time, that PGF_2α, in the form of dinoprost tromethamine has inhibitory effects on growth of M. bovis. Moreover, the effect of PGF_2α on the bacterial growth was not consistent across strains. Nevertheless, if PGF_2α were to be considered as a potential treatment for M. bovis, then a dose for final concentration and administration routes remains to be determined, using in vivo experiments. The use of intra-mammary treatment of mycoplasma mastitis is questionable and needs further research [18].

The antibacterial effects of fatty acids on pathogens have been studied for years and reviewed [9-12]. Our previous results also showed that [19] PGF_2α has inhibitory effects on growth of S. aureus and Streptococcus uberis in vitro [14,15]. In addition, growth of S. aureus and Streptococcus agalactiae was inhibited by certain long-chain fatty acids in vitro [12].

The mechanism by which PGF_2α affected the growth of M. bovis cannot be determined from the current study. Based on current evidence in the scientific literature, we postulate that one potential mechanism of action centers on the ability of fatty acids to penetrate the plasma membrane of bacteria, increasing the negative charge of the bacterial membrane surface, and ultimately disrupting cell membrane integrity [20]. Another possibility involves the hindering of bacterial growth via an interaction of the fatty acids at the cell membrane, resulting in a change in membrane permeability [21] or the disruption of transduction cascades leading to cell lysis [22]. Additionally, since PGF_2α retains some of its hydrophobic properties from its precursor arachidonic acid, it is possible that it is being incorporated into the plasma membrane and may be interfering with membrane fluidity and cellular signaling and transduction cascades similar to mechanisms previously suggested above [21,22].

Mycoplasma are unable to synthesize cholesterol needed to regulate membrane fluidity and must obtain sterols (cholesterol) from their environment to maintain proper fluidity [23,24]. When M. laidlawii were treated with palmitic and steric acids, cell permeability changed, cells appeared irregular, osmotic fragility increased, and cell growth was inhibited [25].

The PGF_2α concentrations used in the current in vitro study were relatively high, and therefore its feasibility as clinical therapeutic agent warrants further investigation. Previous research [12] has shown that linoleic acid and arachidonic acid, precursors of PGF_2α, are potent bacteriostatic fatty acids and at low doses of these fatty acids can inhibit the growth of several gram positive bacteria. The results of the current research provide an opportunity to examine the effect of these and other fatty acids on M. bovis, which currently has no effective treatment.

These in vitro results provide evidence, for the first time, that PGF_2α, in the form of dinoprost tromethamine, has inhibitory effects on the growth of M. bovis, and the bacteriostatic effect appears to be strain and dose dependent. The clinical application of PGF_2α and its efficacy for treatment of M. bovis requires further investigation. These findings provide further research opportunities to investigate the effect of fatty acids on M. bovis.

This study was made possible by the support of Zoetis Animal Health, Florham Park, NJ, in providing pure prostaglandin F_2α (dinoprost tromethamine), Idaho Dairymen’s Association, NIH P20 RR15587, and by the Idaho Agricultural Experiment Station. The authors are thankful to Mr. Ben Enger for the careful review of the manuscript.