DOI: 10.3390/vetsci13101003 ISSN: 2306-7381

Maternal Metabolic Profiles, Their Relationships with Progesterone and Pregnancy-Associated Glycoprotein (PAG), and Biomarker Potential During Early Pregnancy in Lactating Ewes

Gökhan Uyanık, Ahmet Gözer, Murat Abay, Ufuk Kaya, Ishak Gökcek, Ramazan Sertkol

This study aimed to characterize the longitudinal maternal biochemical phenotype associated with early pregnancy, determine its relationships with progesterone (P4) and pregnancy-associated glycoproteins (PAGs), and conduct an exploratory evaluation of the prognostic and discriminatory potential of selected variables. Following synchronized breeding outside the breeding season, serum samples were collected from lactating ewes with confirmed singleton pregnancies (n = 22) and non-pregnant ewes (n = 22) at six time points from mating to D50. Pregnant ewes showed lower total protein and globulin, higher albumin-to-globulin ratio (A/G), and higher lactate dehydrogenase (LDH) from D7 to D35. Total oxidant status (TOS) was lower at mating and D14 in ewes that subsequently became pregnant. In pooled analyses, the P4-PAG correlation increased from D21 (r = 0.353) to D50 (r = 0.844), whereas most biochemical associations were attenuated in pregnancy-only analyses, suggesting that some pooled relationships were partly influenced by pregnancy-status separation. TOS showed the highest apparent prognostic performance at mating (AUC = 0.939). On D7, P4 (AUC = 0.948) and LDH (AUC = 0.915) showed the highest apparent individual discriminatory performance. By D21, P4 provided complete in-sample discrimination (AUC = 1.000), while A/G was the highest apparent performance among non-hormonal variables (AUC = 0.916). Exploratory multivariable models indicated the greatest apparent potential for biochemical–endocrine complementarity during D7-D14. These findings support a dynamic maternal biochemical phenotype and identify routinely measurable candidate variables that may complement luteal and conceptus–placental indicators, although these performance estimates are hypothesis-generating and require external validation.