Physiological and Pathological Alterations in the Mammary Gland of Holstein Cows during Transition Period

Background: Physiological and immunological adaptations occurred in peripartum period may be responsible for susceptibility to intramammary infections during colostrogenesis. Increase in blood flow and hormones cause physiological edema by accumulation of lymphatic fluid in interstitial space. Specific composition of mammary secretion increases the susceptibility to mastitis. So, the study of physical, cellular and microbiological changes in the mammary gland in this period are fundamental for understanding, diagnosing and adopting prophylactic measures for bovine mastitis. This study aimed to characterize physiological and pathological changes that occur in the mammary gland during transition period. Materials, Methods & Results: Thirteen cows, in second to fourth lactation were followed from 3 weeks pre-calving to 3 weeks post-calving. Cows were submitted weekly to physical examination of mammary gland, strip cup test, California Mastitis Test (CMT), Microscopic Somatic Cell Count (MSCC), and bacteriological examination of mammary secretion. Mammary quarters were divided into two groups: BAC: samples that did not present bacteriological growth throughout transition period; BAC: samples that presented bacteriological growth in at least two isolates of the same bacterial genus. Regarding physical examination of mammary gland, the parameters localized or generalized alteration in volume, asymmetry, reddish color, reduced skin elasticity, increased temperature, and swollen consistency presented higher frequencies in the BAC group. On the other hand, infection in mammary quarters of cows from BAC group promoted a decrease in frequencies of increased volume, decreased elasticity, and elevated temperature in the mammary gland, in addition to a higher positivity to CMT and higher cellularity. Overall rate of intramammary infection was 25.82%, with non-aureus Staphylococci (NAS) being the predominant agent (69.2%). Maximum peak of bacterial isolates occurred immediately after calving (33.3%). Differences were observed between groups (P = 0.021) for CMT, showing a higher frequency of positive tests in BAC group (36.5%) compared to BAC group (18%). MSCC varied between weeks (P = 0.001), groups (P = 0.001), and between the interaction of the two variables (P = 0.019). Highest median value of MSCC was observed at calving (1.5 × 10 cells/mL), being different (P = 0.001) from the +1, +2, and +3 weeks and the lowest median value of MSCC (0.08×10 cells/mL) was observed in +3 compared to previous weeks (P < 0.05). Discussion: NAS was predominantly isolated in mammary secretions of cows during peripartum, it is commonly found in the skin and canal of the teat and are associated with subclinical mastitis. Peak of mammary infection was observed on calving, probably by the cortisol peak resulting in a stress leucogram and decrease in the functional activity of immune cells. Changes observed in the physical examination of mammary gland of cows from BAC group are compatible with physiological edema caused by circulatory changes that occurs with the proximity of calving, increasing blood flow necessary for production of colostrum. Alterations in the mammary quarters of cows from BAC group may be associated with functional loss of mammary parenchyma or persistent infectious process, characterizing clinical findings of pathological changes during transition period. Mammary quarters of the BAC group presented higher percentages of positive reactions to CMT in weeks +1 and +2, and higher MSCC values and are associated with infection of the mammary gland.


INTRODUCTION
Mammary gland of cows can undergo successive renovations during cycles that comprise lactation: lactogenesis I (growth, differentiation and colostrogenesis), lactogenesis II (full lactation) and involution (non-lactating state). In lactogenesis I gestation induces changes in endocrine system responsible for proliferation of the mammary parenchyma [14] and IgG1 are transported by mammary epithelial cells through transcytosis four weeks before calving until calving [2].
Increase in blood flow and action of hormones involved in lactogenesis I cause accumulation of lymphatic fluid in interstitial space, promoting physiological edema in udder, making it more susceptible to physical damage [12].
Composition of colostrum and transitional milk are also unfavorable to defense system of mammary gland during peripartum, increasing the susceptibility to mastitis. Despite large number of somatic cells present in these secretions, there is a low proportion and reduced viability of neutrophils [4]. Predominant IgG1 in colostrum does not contribute to phagocytosis of bacteria by polymorphonuclear cells due to its low opsonizing activity [5].
Colostrogenesis may be a risk phase for mastitis in immediate postpartum and it is extremely difficult to differentiate physiological from pathological edema of udder, due to the similarity of clinical signs and inviability of tests that estimate the cellularity of mammary secretion in this period [17]. The aim of this study was to characterize physiological and pathological alterations in mammary gland of Holstein cows during transition period.

Animals and conditions of study
The study was conducted between September and November on an experimental farm belonging to the Paulista Agency of Agribusiness Technology (APTA) located in the city of Nova Odessa -São Paulo, Brazil (latitude 22°75′S and longitude 47°27′W). Thirteen Holstein cows were selected, between second and fourth lactation, according to the date of insemination and calving prediction. Cows were monitored weekly at seven different times during the transition period, including calving (week -3 to week +3), and were submitted to physical examination of the mammary gland (n= 52 mammary quarters, 364 evaluations, weeks -3 to +3), examination of mammary secretion that included strip cup test (n= 156 samples, weeks +1 to +3), California Mastitis Test -CMT (n= 156 samples, weeks +1 to +3), microscopic somatic cell counts -MSCC (n= 208 samples, from calving to +3), and bacteriological examination (n= 364 samples, weeks -3 to +3).
Management procedures adopted during study were according to farm routine, including dry cow therapy using a product (Mastijet fort) 1 containing the antimicrobials tetracycline (200 mg), neominine (250 mg), and bacitracin (2000 IU) with 10 mg of prednisolone.
Cows were transferred to the maternity pen composed of Cynodon grass, 30 days before the expected calving. Regarding the welfare conditions, the absence of coverage was verified; except for the existence of some trees, there were no specific areas for parturition and the deliveries were not attended at night. After calving, animals were transferred to paddocks for newly calved cows and separated in batches according to dairy production.
Composition of the diet offered to cows during pre and postpartum was the same; however, the amounts of corn silage (prepartum: 20 kg, postpartum: 35 kg) and concentrate (prepartum: total equivalent to 3 kg, postpartum: 1 kg for each liter of milk produced) were different. The anionic balance of diet was not considered due to low rates of occurrence of hypocalcemia and retained placenta. Drench postpartum was also not provided.
Physical examination of the mammary gland was performed in a containment trunk, according to Grunert [8]. All mammary quarters were examined by inspection and palpation. Parameters used in the physical examination are presented in Table 1.

Samples and parameters
After physical examination, the cows were transferred to milking parlor in order to ensure the hygiene of the samples for bacteriological examination. Teats were predipped with 2% sodium hypochlorite solution and dried with individual paper towels; then the first mammary secretion jets were discarded for the strip cup test.
For bacteriological examination, 3 mL of mammary secretion was obtained in sterile plastic bottles with screw caps, after disinfecting teats with 70% alcohol [9] and it was performed according NMC [14].
Samples were considered to have an intramammary infection (IMI) if they had three or more colonies of the same type. Samples with three or more species on a plate were considered contaminated. If no bacterial growth was observed after 72 h on the initial culture the incubated sample was replated.
California Mastitis Test (CMT) and Microscopic Somatic Cell Count (MSCC) were performed from parturition. For CMT, samples with negative or traces results were considered negative because of the high cellularity expected for this phase of lactation, whereas 1+ to 3+ reactions were considered positive [21]. For MSCC, 50 mL of mammary secretion was obtained in conical plastic tubes [6].
Mammary quarters that did not present bacterial isolation at any of the study times (-3 to +3) were included in the BACgroup, while those with at least two isolates of the same bacterial genus were grouped in the BAC + group.

Statistical analysis
Statistical analysis was performed using the Statistical Social Package Science program -version 18. Variables of the study were qualitative (binary and ordinal) and quantitative. Generalized estimating equations (GEE) from a generalized linear model (GLM) was used with a logistic model for qualitative binary and ordinal variables. The quantitative MSCC variable was tested for the type of distribution and adjusted for the best model according to the Akaike information criterion (AIC). For the latter, the model of distribution Gama on the GEE was adopted. Time and group were the predictor variables tested for effect and interaction. In the definition of the model, exponential parameters were proposed to obtain odds ratio (OR) and confidence interval.

RESULTS
During the study, some samples were discarded due to insufficient volume and manifestation of disease (ketosis) in three cows. In addition, during the sample division into the experimental groups (BAC + and BAC -), eight mammary quarters needed to be excluded from the study because they had only one bacterial isolation, making it impossible to perform appropriate statistical tests. Thus, 302 samples were evaluated for bacteriological examination (151 samples in the BAC + group and 151 samples in the BACgroup, between weeks -3 to +3), 303 physical evaluations of the mammary quar-ters (152 in the BAC + group and 151 in the BACgroup, between weeks -3 to +3), 128 samples for CMT (63 in the BAC + group and 65 in the BACgroup, between weeks +1 to +3), and 174 samples for MSCC (86 in the BAC + group and 88 in the BACgroup, between the week of calving to the week +3).
The effect of the group (BAC + and BAC -), evaluated weeks (-3 to +3), and interaction between groups and weeks on the physical parameters of the mammary gland, besides the odds ratio are presented in Table 1. The parameters generalized volume alteration, elasticity of the skin, and increase in temperature of the mammary gland differed between the experimental groups (P = 0.001; 0.011 and 0.003, respectively), with the highest frequencies occurring in the BAC -. Evaluation of the effect of time (weeks) on the physical parameters revealed higher frequencies in the BACgroup for generalized volume alteration (P = 0.001), localized volume alteration (P = 0.044), asymmetry (P = 0.001), reddish coloration of the mammary gland (P = 0.019), decreased elasticity (P = 0.001), increased temperature (P = 0.001), swollen consistency of the mammary gland (P = 0.001), and altered consistency of the teats (P = 0.001). No significant differences were observed when the interaction between groups and time was evaluated. The strip cup test was positive for only one mammary quarter at weeks +1 and +2 (BAC + group), setting as the only positive samples for clinical mastitis.
According CMT, differences were observed between groups (P = 0.021), with a higher frequency of positive tests in the BAC + (36.5%, 23/63) compared to BAC -(18%, 12/65). BAC + group had a greater chance of presenting a positive result during transition period (OR = 2,585, 95% CI = 1,153-5,797). However, time and interaction between time and groups were not significant for CMT (P = 0.560 and 0.640, respectively).

DISCUSSION
This study evaluated physiological and pathological changes occurred in the mammary gland and milk examination of Holstein cows during transition period. To our knowledge, there are no published studies addressing the differentiation between physiological and pathological edema of the mammary gland, which directly aids in the diagnosis of clinical mastitis.
Overall rate of mammary infection for the transitional period was 25.82%. NAS were predominantly isolated in the mammary secretion of cows during peripartum (69.2%), similar to the frequency of 76.92% observed by Reis et al. [17] in the same property selected for the present research, while Odenstsen et al. [15] found frequencies of 31.7%, indicating that sanitary management of the property directly impacts the rates and type of bacterial agents that cause mammary infections.
NAS bacteria are commonly found in the skin and canal of the teat and are associated with subclinical mastitis [20]; this aspect was verified in the present study, since of the 78 positive samples in the bacteriological examination, only one presented a positive result in the strip cup test. A previous study showed that some factors are associated with NAS infections, such as primiparous cows and a history of high SCC in cows [16].
Cows evaluated in the present study were treated during the dry period with a commercial product containing antibiotics and corticoid 1 . However, although dry cow therapy was used to prevent new occurrence of intramammary infection between the dry period and subsequent onset of lactation [23], this was not sufficient to avoid bacterial growth occurring during the transition period. The antimicrobials present in the formulation of commercial product are broad-spectrum, but probable failures in milking management and resistance to antimicrobials may have compromised the effectiveness of dry cow therapy [19].
The peak of mammary infection was observed on the day of parturition (33.3%); the cortisol peak occurred due to the stress of calving, combined with the negative energy balance of the transition period, resulting in a stress leucogram and probable decrease in the functional activity of immune cells [7]. In addition to changes in immunity, physiological changes in the mammary gland itself such as increased volume, increased skin tension, and pendulous udders are risk factors for intramammary infection [12].
Circulatory changes occur with the proximity of calving, increasing blood flow to the mammary gland necessary for the production of colostrum. This event occurs weeks before calving and ceases abruptly on calving under hormonal influences [1]. These phenomena justify the changes observed in the physical examination of the mammary gland of cows belonging to BACgroup, such as generalized or localized increase in mammary gland volume, asymmetry, reddish coloration, decrease in skin elasticity, increase in temperature, and swollen consistency, which are compatible with the postpartum mammary gland physiological edema observed in 95% of the cows [10].
The intensity of the physiological edema of the mammary gland can vary according to the severity of the changes, and can be classified in scales varying from 1 to 5 [3] or from 1 to 10 [22]. The presence of positive Godet signal is observed in edema, which mainly accompanies enlargement of the mammary gland, decrease in skin elasticity, and increase in temperature; these parameters were altered during the study period for the BACgroup.
Mammary edema may also vary according to number of lactations, race, and season of the year; however, few studies have demonstrated changes in physical parameters during transition period, in terms of the differentiation between physiological and pathological edema [3,10].
On the other hand, infection in the mammary quarters of cows belonging to the BAC + group promoted a decrease in the frequencies of increased volume, decreased elasticity, and elevated temperature of the mammary gland and may be associated with functional loss of the mammary parenchyma or persistent infectious process, characterizing the clinical findings of pathological changes in the mammary gland during the transition period.
Some studies have demonstrated the impact of infection on the health of mammary gland, such as changes in physical characteristics of the milk, damage to the mammary epithelial cells, and gene expression in these tissues [9,18]. Fundamental in the regulation of immune response, endothelial cells of mammary gland exert major changes in episodes of mammary infection, facilitating the passage of leukocytes and changing blood flow, factors that are directly associated with loss of mammary tissue [18]. Other impacts such as expression of genes in cases of infection of mammary gland have been studied, indicating that levels of expression depend on the duration of infection and type of bacteria [9].
Mammary quarters of the BAC + group presented higher percentages of positive reactions to CMT in weeks +1 and +2, in addition to higher MSCC values. These milk test results show that the increases in these parameters are associated with the infection of the mammary gland, indicating differences between groups.
The result obtained for MSCC of colostrum in the present study (1.5 × 10 6 cells / mL) differed from the values of 0.88 × 10 6 cells / mL observed by Gomes et al. [6]. The status of mammary healthiness and bacterial infection rate were responsible for the observed differences.
MSCC was higher in the colostrum compared to weeks +1, +2, and +3. The decrease in the MSCC in the first week postpartum has also been reported by Mehrzad et al. [11]. In the present study, low milk cellularity between weeks +1 to +3 is compatible with the low rate of mammary infection observed in the respective weeks. Accumulation of leukocytes during the colostrogenesis process and desquamation of the epithelial lining cells of mammary gland may be responsible for the high cellularity of colostrum. However, as milking progresses, the elimination of these cells, bacteria, and cellular debris accumulated in the mammary secretion and a subsequent decrease in milk cellularity occur. Physiological changes in cellularity in the immediate postpartum compromise the use of the MSCC for the detection of mastitis [17].
In general, the BAC + group presented a lower percentage in the frequencies of increased volume, decrease elasticity, elevated temperature of the mammary gland, and higher value of CMT, MSCC, and bacterial infection rate compared to the BACgroup. These phenomena are probably related to the histological characteristics of the mammary tissue, associated with the chronic inflammatory process in the group of cows with mammary infection manifested during the transition period.

CONCLUSIONS
In this study we characterized physiological and pathological alterations in mammary gland and we observed during the physical examination higher frequencies of parameters such as generalized or localized alteration in volume, asymmetry, reddish coloration, decreased elasticity of the skin, increased temperature and swollen consistency in cows free of mammary infection and carriers of postpartum physiological edema. In the BAC + group, a lower frequency was observed for changes in volume, elasticity, and temperature, higher frequencies of samples positive for CMT test and higher cellularity, which are characteristics compatible with pathological alterations. This study contributes to the diagnosis of mastitis during the transition period through physical examination of the mammary gland and evaluation of its secretion.

Declaration of interest.
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of paper.