Timing of pregnancy losses in grazing lactating dairy cows based on cervical interferon-stimulated gene-15 mRNA, pregnancy-associated glycoproteins, and progesterone dynamics: An exploratory study.

Pregnancy loss (PL) in lactating dairy cows disrupts reproductive efficiency, yet the timing and physiological mechanisms underlying early losses remain poorly defined. We evaluated cervical interferon-stimulated gene (ISG) expression (d 12, 14, and 16 after artificial insemination [AI] in inseminated cows; corresponding days of the estrous cycle in cyclic controls), circulating progesterone (P4; d 12, 14, 16, and 21 after AI or equivalent cycle days), and pregnancy-associated glycoprotein (PAG; d 28 after AI) profiles in lactating Holstein cows (n = 67; 57 inseminated and 10 cyclic controls) to characterize temporal patterns associated with pregnancy establishment and loss and to infer the timing of pregnancy loss. Pregnancy was also diagnosed by transrectal ultrasonography on d 28 and 60. Based on individual longitudinal ISG profiles, evidence of conceptus signaling (≥2.5-fold increase in d 12, 14, or 16, alone or in combination) was detected in 94.7% of inseminated cows, indicating 5.3% loss before d 12 (nonfertilization or early embryo degeneration). Subsequent PL occurred in 9.3% (5/54) of cows between d 12 and 16 (elevated ISG15 on d 12 and 14 but not on d 16), in 32.7% (16/49) between d 16 and 21 (elevated ISG15 on d 16 but low P4 by 21), in 24.2% (8/33) between d 21 and 28 (elevated ISG15 d 16, elevated P4 d 21, but no embryo or PAG on d 28), and in 20.0% (5/25) between d 28 and 60 (embryo and PAG d 28 but no embryo d 60), resulting in pregnancy per AI of 43.8% at d 28 and 35.1% at d 60. For cows with PL from d 12 to d 16 or PL d 16 to 21, corpus luteum regression was detectable by d 21 (low P4, ≤1 ng/mL). In contrast, cows with PL between d 21 and 28 exhibited ISG and P4 profiles similar to pregnant cows through d 16 to 21 but had lower PAG concentrations on d 28. Cows with PL between d 28 and 60 had PAG concentrations consistent with pregnancy at d 28, indicating conceptus attachment followed by subsequent pregnancy failure. An exploratory evaluation of other ISG in the cervical cytobrush collections on d 12, 14, and 16 in a subset of cows (n = 32) indicated that they provided complementary information to ISG15, although IFI44 enabled detection of pregnant cows at the earliest time point (d 12). In conclusion, PL before d 21 was associated with patterns consistent with reduced conceptus IFN signaling and insufficient luteal support, whereas PL after d 21 occurred despite evidence of early conceptus signaling and functional luteal maintenance, suggesting a shift from failure of pregnancy recognition to failure of pregnancy maintenance. Cervical ISG profiling may provide a useful approach to determine the timing and underlying biology of PL in dairy cows.