Trans-rectal ultrasound examination of all the twenty inseminated repeat breeders on 26
th day post-insemination revealed the presence of amniotic vesicle in 9 (45%) cows and non-pregnancy in 11 (55%) cows. The finding of the present study is similar to the previous reports on use of ovsynch protocol to improve pregnancy rate in repeat breeders
(Fricke et al., 2003; Caraba and Velicevici, 2013;
Jayaganthan et al., 2016).
Days of the animals in milk for pregnant and non-pregnant animals were 236±24 and 337±50 respectively. On comparing the serum cortisol level of two groups, significant increase was observed on 18
th (P<0.0001), 20
th (P<0.0001), 22
nd (P=0.0004) and 25
th (P=0.0003) days of post-insemination in non-pregnant animals (Fig 2a). Significant increase in salivary cortisol was observed on 18
th (P<0.0001), 20
th (P<0.0001), 22
nd (P<0.0001) and 25
th (P=0.0130) days of post-insemination in non-pregnant animals when compared with pregnant animals (Fig 2b). On comparing with pregnant animals, urinary cortisol level of non-pregnant animals was significantly increased on 18
th (P<0.0001) and 20
th (P=0.0008) days of post-insemination (Fig 2c). These significant increase in serum, salivary and urinary cortisol level of non-pregnant animals compared to that of pregnant animals, indicates the association of cortisol and early pregnancy complications
(Thatcher et al., 1984; Schafer-Somi, 2003). Previous studies in repeat-breeding cows on early embryonic mortality have also recorded loss of embryos in the range from 8% to 30%
(Boyd et al., 1969; Hawk, 1979;
Ayalon, 1978;
Thatcher et al., 1994) with most embryonic losses occurring between 8 and 16 days of breeding
(Roche et al., 1981). Initial trophoblast-decidua interactions occur through molecular dialogue initiated during blastocyst attachment by signalling molecules like integrins and fibronectin
(Burrows et al., 1996) but physiological concentration of glucocorticoids can suppress the expression of these signalling molecules
(Ryu et al., 1999). Furthermore, successful implantation requires co-ordinated role of pro-inflammatory cytokines, such as interferones, interleukins-1,2,4,6,10, granulocyte macrophage -colony stimulating factor, tumor necrosis factor-α (
Schafer-Somi, 2003;
Raheem, 2018) but glucocorticoids were known to impair this cytokine signalling cascade required for implantation (
Michael and Papageorghiou, 2008).
Comparing serum cortisol of non-pregnant animals with different post-insemination days, revealed a significant increase on 18
th and 20
th days than 14
th (18
th day: P=0.0157; 20
th day: P=0.0273) and 22
nd (20
th day: P=0.0253) days (Fig 3b). However, no significant difference was observed in the serum cortisol of pregnant animals among different post-insemination days (Fig 3a).
Within group comparison of salivary cortisol in non-pregnant animals revealed significant increase on 18
th and 20
th days than 14
th (18
th day: P=0.0001; 20
th day: P=0.0009), 22
nd (18
th day: P=0.0033; 20
th day: P=0.0089) and 25
th (18
th day: P=0.0013; 20
th day: P=0.0022) days (Fig 4b). No significant change was seen in salivary cortisol of pregnant animals on different post-insemination days (Fig 4a). Among different post-insemination days the urinary cortisol level of non-pregnant animals showed a significant increase on 18
th and 20
th days than 14
th (18
th day: P=0.0016; 20
th day: P=0.0143) and 22
nd (18
th day: P=0.0027; 20
th day: P=0.0125) days (Fig 5b). Urinary cortisol level of pregnant animals did not differ significantly on different days of post-insemination (Fig 5a).
Significant variation in the cortisol level among different post-insemination day was observed only in non-pregnant animals. Reports of previous studies suggests that cortisol is an important factor in inducing the pulsatile release of PGF2α from the inter caruncular region of deciduas, which causes the regression of corpus luteum and reduces the circulating level of progesterone needed for pregnancy maintenance
(Ginther et al., 2009; Duong et al., 2012; Duong et al., 2012a). Cortisol can also indirectly suppress the progesterone level by reducing basal level of luteinizing hormone as well as luteinizing hormone - releasing hormone induced release of luteinizing hormone from pituitary cells of bovine
(Padmanabhan et al., 1983).
In the current study, salivary and urinary cortisol levels of non- pregnant animals were found to be positively correlated with serum cortisol in all the days of study (Table 1), which indicates salivary and urinary cortisol can be used as an indicator for measuring the hypothalamic-pituitary-adrenal axis activity in repeat breeding cows. Plasma cortisol has been routinely used as an indicator of stress in animals (
Broom and Johnston, 1993) but it provides only single point-in-time estimate and also restraining of animals during blood collection may interact with plasma cortisol level (
Willett and Erb, 1972;
Alam and Dobson, 1986). Plasma cortisol measures the total cortisol which includes both corticosteroid-binding globulin bound cortisol and free cortisol
(Coolens et al., 1987). Alternatively, salivary and urinary samples can be used to estimate free cortisol. Salivary cortisol concentration was unaffected by salivary flow rate because cortisol enters saliva by passive diffusion
(Riad-Fahmy et al., 1982) and also it was highly correlated to changes in serum cortisol level
(Vining et al., 1983). Urinary cortisol reflects systemic cortisol concentration that enters the urine
via glomerular filtration
(Biesel et al., 1964) and also it has the merit of providing an integrated index of cortisol secretion over the period preceding sampling (
Lasley, 1985). These non-invasive cortisol levels observed in the present study correlate with serum cortisol in various species including horses
(Peeters et al., 2011), cattle
(Perez et al., 2004), rats
(Thanos et al., 2009) and humans
(Hellhammer et al., 2009).
Serum cortisol level was positively associated with pregnancy rate on day 18
th (P<0.0001), 20
th (P<0.0001), 22
nd (P<0.0001) and 25
th (P<0.001) days of post insemination and salivary cortisol level was positively associated with pregnancy rate on day 18
th (P<0.0001), 20
th (P<0.0001), 22
nd (P<0.001) and 25
th (P<0.001) days of post insemination. Urinary cortisol level was positively associated with the pregnancy rate on 18
th (P<0.0001), 20
th (P<0.05) and 22
nd (P<0.05) days of post insemination (Table 2). This indicates that the systemic cortisol level may be used to predict the glucocorticoid induced early pregnancy loss in repeat breeders. Previous studies have reported that in repeat breeders the developmental ability of embryo was compromised (
Linares, 1981,
Gustafsson and Larsson, 1985;
Albihn et al., 1989), as cortisol can induce the luteolysis and reduce the circulating level of progesterone (
Spicer and Chamberlain 1998). Depending upon the physiological status, cortisol can alter the embryo implantation and pregnancy rate in heifers
(Duong et al., 2012). Elevation of cortisol concentration is associated with induction of oocyte apoptosis and deteriorates oocyte quality
(Prasad et al., 2016), percentage of abnormal oocytes in repeat breeders may range up to 52.5%
(Kurykin et al., 2011). Chronic and acute stress can affect pulsatile release reproductive hormones from hypothalamic-pituitary-gonadal axis and reduces the reproductive performance of animals
(Walker et al., 2008; Dobson et al., 2003; Smith et al., 2003). Hence, cortisol can be used as a biomarker for assessing the stress induced early pregnancy changes in repeat breeders.