Mastitis is one of the major problems related to milk production, because it causes economic losses, resulting from the reduction in volume of milk produced, and of the costs of treatment. Sometimes there is even animal loss. Although well known in bovine species, studies about this disease in ovine species are scarce (^{BAGLITZ et al., 2008}).

The diagnosis of subclinical mastitis can be performed through the somatic cell count (SCC), i.e., the examination of structures of the organism defense before inflammation or infection and composed by immune system cells (somatic cells) and epithelial cells (^{SUMMER et al., 2012}). As an indicator of the mammary gland health and the milk quality, the SCC is quite studied in bovine species (^{VIANA et al., 2010}). However, the ewe’s milk is an apocrine secretion and presents cytoplasmic particles, which have similar size and shape to the leucocytes, and that can generate some confusion in automated methods of cell count (^{MADUREIRA et al., 2010}). Therefore, the count of cells by direct microscopy is indicated for the ovine species. However, there is disagreement regarding the staining procedures suitable for this species (^{GOMES et al., 2008}), and the protocols of specific staining for DNA are the ones indicated.

Considering the laborious processes for specific staining for DNA, as the color with methyl green and pyronin Y, as well as its high carcinogenic potential (^{BAGLITZ et al., 2013}), the SCC by non-specific staining for DNA can be an applicable tool for the diagnosis of mastitis in sheep (BERRY et al., 2010; ^{GOMES et al., 2008}). In this perspective, the objective of this study was to identify a methodology of fixation and staining for SCC by optical microscopy in ewes’ milk.

The investigation was carried out in a commercial herd for milk production, with dairy sheep in different phases and number of lactations, located in the serrana region of Rio Grande do Sul state, Brazil. Samples were collected from individualized milk by teat, in sterile flasks, and identified, after asepsis with 70% alcohol and disregarding the first three jets. The samples were packed in isothermal boxes with ice packs for later processing of different methods of fixation and staining.

For the slides preparation, an aliquot of 10 µL of sample was deposited on the glass slide, supported on mold with an area of 1 cm² (^{SCHALM et al., 1971}). The slides were kept at room temperature for 24 hours, for drying. They were subsequently fixed and stained, creating six different combinations:

For each combination, 48 slides were prepared.

In the first procedure for the milk film fixation, the sample was exposed to heat in a Bunsen burner and steeped in xylene for 10 minutes, for the removal of fat, and washed with distilled water buffered with pH 7.2. Hereinafter, the slide was submerged in ethyl alcohol for 20 minutes and washed with distilled buffered water, at pH 7.2.

In the second procedure, the samples were submitted to fixation with Carnoy’s solution - composed of ethanol (60%), formaldehyde (30%), and acetic acid (10%) (^{PUCHTLER et al., 1968}) - for ten minutes (^{GOMES et al., 2010}). After, they were hydrated for 1 minute with alcohol 50% followed by alcohol 30%, for more 1 minute. Then, the slides were washed in distilled water for 1 minute (^{ARCURI et al., 2004}).

Subsequently, the fixed slides were subjected to one of the four staining procedures: May-Grünwald-Giemsa (Laborclin Produtos para Laboratórios, Pinhais, Paraná, Brazil); Wright (Laborclin Produtos para Laboratórios, Pinhais, Paraná, Brazil); and Panótico (Laborclin Produtos para Laboratórios, Pinhais, Paraná, Brazil), according to the manufacturer’s instructions, and Broadhurst-Paley as described by ^{SCHALM et al. (1971}).

The obtained slides were observed regarding characteristics of smear, fixing and in optical microscope, with immersion oil in 100 × magnification. Microscopy characteristics and cells general appearance were observed.

The characteristics presented in different combinations of fixers/stainings were used to describe the appearance of the obtained slides, and seven groups were created, which effectively describe what was observed in the slides:

The characteristics pointed before made the cell count difficult, or even impossible (Fig. 1).

Figure 1: Organization of groups of slides for somatic cell counts in ewes’ milk, according to the presented characteristics. 

The protocol using xylol as fixative and Broadhurst-Paley staining was the one that presented the greatest diversity of characteristics in the slides (blurry, dirty field and little stained), which made the visualization and cell count difficult and impossible, respectively (Table 1). This is the protocol described by ^{SCHALM et al. (1971}) for the determination of SCC in bovine milk, which was also used for the determination of SCC in caprine milk (^{STEHLING et al., 1988}). However, the average fat content of 3.9, 4.5 and 7.2% in cow, goat and sheep milk, respectively (^{MENDONÇA et al., 2010}), should be considered, because the high fat content of ovine milk can make the reading difficult to be performed.

In the present study, the fat content of the milk sheep made the fixation of milk aliquots in the slides a challenging step, since in the four protocols in which xylol was used as a fixative in only 0.52% (1/192 slides) it was possible to perform the SCC.

The rinsing, the last step of the staining, presented itself as a critical point, since the samples, which seemed to be fixed, debonded from the slides and were lost.

In the protocols associated with xylol, in most of the slides loss of the counting area limit the presence of artifacts was observed. The staining of Wright and Panóptico showed no good cell definition, and the milk film was smudged (Table 1). ^{VIANA et al. (2010}), using the Panótico staining, observed no changes in cell morphology of bovine milk.

In the studied protocols, the characteristic 7 enabled cell count, because the cells were stained in a satisfactory way and well fixed to the slide, with a well delimited area and, practically, without artifacts that could harm the counting procedure. This characteristic was observed in most of the slides of the model PAD6 (Table 1). Although Carnoy’s solution has also been used associated to May-Grünwald-Giemsa coloring (PAD5), in this combination most of the slides showed the formation of artifacts that prevented the cell count.

May-Grünwald-Giemsa and Broadhurst-Paley stainings were efficient. However, Broadhurst-Paley staining presented better visualization and the possibility of a differentiation of cell types. In addition, it presented a better distinction of cytoplasmic corpuscles (Fig. 2), avoiding that they could be mistakenly included in the cell count. Although ^{VIANA et al. (2010}), on the other hand, have observed changes in cell morphology in bovine milk with the Broadhurst-Paley staining.

Figure 2: Smear of ovine milk for somatic cell counting prepared with Carnoy’s solution and Broadhust-Paley staining, evidencing: (A) mononuclear cells; (B) polymorphonuclear; (arrows) corpuscles of inclusion (magnification 100 ×). 

SCC by direct microscopy in ovine milk is still little studied and needs greater attention, being an alternative for areas distant from the reference centers with specific equipment for implementation of automated SCC.

Based on the observed results, it was determined that the protocol using Carnoy’s solution and Broadhurst-Paley staining was more effective for the removal of fat and the film adherence to milk glass slides. It had easy implementation, fewer steps and less time for performing the technique.