MATERIAL AND METHODS

MILK SAMPLES COLLECTION

Ninety-one Holstein Friesian dairy herds from Los Rios (66) and Los Lagos (25), the two most important dairy Regions in the country, were selected by convenience. The herds had between 50 and 400 lactating cows and were providers of the "Sociedad Procesadora de Leche del Sur S.A" (PROLESUR). Most of them were managed under grazing-based systems with concentrate and silage as winter supplementation. Individual milk samples were collected from each bulk milk tank by trained personnel from the PROLESUR company in accordance with the recommendations of the National Mastitis Council, USA (^{Hogan et al 1999}). All the samples were kept and transported at 4 °C to the laboratory within 8 h of collection for bacteriological examination. Additionally, 100 individual milk samples from cows with >500,000 cells/mL and negative bacteriological results for traditional mastitis pathogens were included. All these milk samples were received in our diagnostic laboratory and came from a dairy herd located in the Biobío Region, with a high incidence of clinical mastitis refractory to antibiotic therapy.

BACTERIOLOGICAL ANALYSIS

Immediately after arrival at the laboratory, 100 pL of each milk sample were plated onto the surface of PPLO Medium (Difco), supplemented with yeast extract, horse serum, salmon DNA, thallium acetate and penicillin and incubated at 37 °C with 10% CO₂ (^{Hogan et al 1999}). The plates were read at 3, 7, 10 and 12 days of incubation using a stereoscopic microscope (40x), observing all the streaks made on the agar to detect Mycoplasma colonies. At the end of the incubation period, the plates with no growth were discarded and those growing small colonies with a "fried egg" appearance were considered suspicious of Mycoplasma. Then, the surrounding agar of the suspicious colonies was cut out and subcultured into PPLO broth supplemented as mentioned above, and incubated for 3 days at 37 °C with 10% CO₂, to obtain a pure culture. M. bovis ATCC 25025 strain was used as a positive control for all the tests.

For DNA extraction, 5 mL of a pure culture was centrifuged at 12,000 g for 30 seconds, the supernatant discarded, and the pellet resuspended in 150 pL of PBS, and processed with the commercial kit "AxyPrep Bacterial Genomic DNA Mini-Preparation Kit" (Axygen, Inc.) following the manufacturer instructions.

GENUS IDENTIFICATION

For the Genus differentiation between Mycoplasma and Acholeplasma, a common environmental organism, the digitonin disk test and a specific PCR test were used (^{Boonyayatra et al 2012}). The digitonin disk test was performed using a sterile cotton swab moistened with pure culture broth of the suspicious colony, which was inoculated onto the entire surface of a PPLO agar plate. Afterwards, a digitonin disk (Udder Health System, Inc) was placed in the centre of the plate and incubated for 4 days at 37 °C with 10% CO₂. The presence of a growth inhibition zone surrounding the digitonin disk was visually checked. A clear inhibition zone >5 mm from the edge of the disk was considered as a positive test for the genus Mycoplasma. On the contrary, when the growth inhibition zone was <3 mm it was considered as a negative test indicating that the strain belongs to the genus Acholeplasma. In addition, a specific PCR test for genus differentiation was performed using the set of primers F2 5'-GTG(C/G)GG(A/C) TGGATCACCTCCT-3' and R2 5'-GCATCCACCA(A/T) A(A/T)AC(C/T)CTT-3' which targets the 16S-23S rRNA intergenic spacer region of Mycoplasma and R34 5'-CCACTGTGTGCCCTTTGTTCCT-3' of the 16S-23S rRNA intergenic spacer region of Acholeplasma. (Boonyayatra et al 2012). The amplification program was an initial denaturation cycle of 5 minutes at 94 °C, followed by 35 cycles of 30 seconds at 94 °C, 2 minutes at 55 °C and 2 minutes at 72 °C, with a final extension of 5 minutes at 72 °C. Amplification products were separated by 2% agarose gel electrophoresis stained with ethidium bromide. The presence of two bands was considered a positive result for Acholeplasma, while Mycoplasma produces only one band.

SPECIES IDENTIFICATION

Once the suspect strains were confirmed as Mycoplasma, a specific PCR was performed for the identification of M. bovis. The primers mb-mp 1F 5'-TATTGGATCAACTGCTGGAT-3 'and mb-mp 1R 5'-AGATGCTCCACTTATCTTAG-3' were used for the amplification of the mb-mp 81 gene, as described by ^{Foddai et al (2005}). The thermocycler was set with an initial denaturation cycle of 5 minutes at 94 °C, followed by 30 cycles of 1 minute at 94 °C, 1 minute at 54 °C and 1 minute at 72 °C, with a final extension cycle of 10 minutes at 72 °C. The amplification product was separated by electrophoresis on 2% agarose gel stained with ethidium bromide. The presence of a band of approximately 447 bp was considered as M. bovis. Additionally, all strains were identified by sequencing the 16S rRNA gene, according to the protocol described by ^{Botes et al (2005}). PCR products of approximately 1048 bp in size were purified using the commercial kit Wizard® SV Gel and PCR Clean-Up System (Promega) according to the manufacturer's instructions. Once the PCR products were purified, they were sequenced bidirectionally in MACROGEN (Korea). Consensus sequences were compared to sequences deposited at GenBank using the NCBI's nucleotide-nucleotide BLAST program. An identity above 98.6% was considered as the same Mycoplasma species (^{Kim et al 2014}).

RESULTS AND DISCUSSION

Mycoplasma was isolated in 3 out of 91 (3.3%) bulk tank milk samples. Typical colonies were 300-400 pm in diameter, with a dense central nucleus, surrounded by a lighter peripheral area of growth, which gives it the characteristic "fried egg" appearance (figure 1). Additionally, Mycoplasma was also isolated in 2 out of 100 (2%) individual milk samples from cows with subclinical mastitis. All colonies were identified as belonging to the Mycoplasma genus by both the digitonin and PCR tests. These results show that Mycoplasma is present in dairy herds of southern Chile, however, it must be considered that its presence may be underestimated due to the low sensitivity of a single bulk tank culture. Despite this disadvantage, the quality of the Mycoplasma bulk milk tank culture can be improved by performing three consecutive cultures separated by 3-4 days (^{Gonzalez and Wilson 2002}).

The three isolated strains from bulk tank milk samples were named J14, J35 and J71. Regarding the species identification, only strain J35 was identified as M. bovis by the specific PCR, which produced a band of approximately 447 bp (data not shown), and by partial sequencing of the 16S rRNA gene. Strains J14 and J71 were identified at the species level as M. bovigenitalium by partial sequencing of the 16S rRNA gene. The two strains isolated from cases of subclinical mastitis were named V49 and V61, both identified as M. alkalescens by partial sequencing of the 16S rRNA gene (table 1).

Figure 1  Mycoplasma colony on PPLO agar plate, with typical "egg fried" appearance (400x). 

In Chile, M. bovis had been isolated in 5 out of 71 (7%) bulk milk tank samples of herds from Los Ríos and Los Lagos Regions, from different dairy farms (^{Sickles et al 2000}). The percentage of Mycoplasma positive milk tanks was higher in this study, but the results are not comparable because in our study we used only samples from suppliers of PROLESUR company. Another Chilean study reports that Mycoplasma was isolated from 4 out of 11 (36%) bulk tank milk samples from the Biobío Region, but the strains isolated were not identified at species level (^{Bustos and Muñoz 2011}). Although the percentage of Mycoplasma-positive bulk tanks found in our study was similar to those reported in other countries (^{Fox 2012}), further studies are needed to find out the real prevalence of Mycoplasma in Chile.

We found only one bulk tank milk positive for M. bovis, which indicates that there is, at least, one cow with Mycoplasma mastitis in the herd, so it would be advisable to identify all infected animals and segregate or remove them from the herd (^{Fox 2012}, ^{Nicholas et al 2016}). M. bovis causes more severe cases of intramammary infections due to Mycoplasma, and has been associated with respiratory infections, otitis, septic arthritis, and reproductive infections in cattle (^{Nicholas and Ayling 2003}, ^{Maunsell et al 2011}). Intramammary infection due to Mycoplasma bovis is feared by dairy farmers and veterinarians because it is highly contagious, hard to control and there are still no vaccines available, as for other udder pathogens (^{Mella et al 2017}).

In addition, this study describes for the first time in Chile the isolation of M. bovigenitalium and M. alkalescens. M. bovigenitalium has been documented as a cause of diseases in dairy herds, mainly associated to genital tract infections (^{Lysnyansky et al 2009}) and cases of clinical mastitis (^{Baumgartner et al 2006}, Lysnyansky et al 2016). It is one of the three most isolated Mycoplasma species from bulk milk tanks, with an isolation frequency that varies between 1% and 25% (^{Fox 2012}). Occasionally, M. bovigenitalium and M. alkalescens can be isolated from milk samples, but not always their presence is associated with disease (Lysnyansky et al 2016). However, in our report M. alkalescens was isolated from cows with subclinical mastitis that had negative cultures results to other common mastitis pathogens, so it is likely to be the causative agent of intramammary infection in these animals.

In conclusion, Mycoplasma mastitis is a potential risk for dairy herds in southern Chile since, although in low percentages, the agents are present in the herd environment. It is important to note that not only M. bovis is present in these dairy herds, but also other pathogenic species not previously described in Chile such as M. bovigenitalium and M. alkalescens. Consequently, diagnostic laboratories should attempt to isolate Mycoplasma spp. when testing milk samples from cows with clinical mastitis that do not respond to antibiotic treatment.