Equine fetal-sex determination from mid-gestation to term using serum conjugated steroid profiling by liquid chromatography-tandem mass spectrometry.

Overview

• This research developed a non-invasive and accurate method for determining the sex of equine fetuses during mid to late pregnancy by analyzing specific steroid hormones in the mare’s serum using liquid chromatography-tandem mass spectrometry (LC-MS/MS).
• The study identified distinct differences in conjugated steroid hormone profiles between male and female pregnancies, enabling reliable fetal-sex prediction beyond current methods.

Introduction and Background

• Fetal sex determination in horses is important for breeding management and can inform care and decision-making during pregnancy.
• Existing methods such as ultrasonography are limited by timing constraints, operator skill, and sometimes lack accuracy, especially as pregnancy advances.
• Conjugated steroids such as estrogen and androgen metabolites circulate in the mare’s blood and reflect fetal-placental endocrine activity, potentially differing between male and female fetuses.
• Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is a highly sensitive and specific analytical technique capable of profiling multiple steroid conjugates simultaneously, making it well suited for this application.

Objectives

• To refine the profiling of conjugated estrogens and androgens throughout equine pregnancy using LC-MS/MS.
• To identify whether maternal serum steroid profiles differ by fetal sex during gestation.
• To develop and validate statistical models for accurate, non-invasive fetal-sex determination using conjugated steroid data.

Methods

• Sample Collection:
  • Serum samples were collected from 141 pregnant mares of various breeds starting at 16 weeks of gestation.
  • Samples were grouped into 28-day gestational blocks, spanning mid-gestation to term (blocks 5-11 approximately corresponding to weeks 16-44).
• Steroid Analysis:
  • Analytes included estrone sulfate (E1S), estradiol sulfates, equilin sulfates, glucuronides, and dehydroepiandrosterone sulfate (DHEA-S).
  • A validated LC-MS/MS assay was used for precise quantification, with rigorous method validation covering:
    • Linearity and dynamic range
    • Trueness and accuracy of measurements
    • Precision and repeatability
    • Quantification limits and sensitivity
    • Recovery rates and matrix effects (impact of serum components)
    • Carryover assessment to prevent contamination
    • Stability of analytes during processing and storage
• Data Analysis:
  • Statistical effects of fetal sex, breed, parity (number of pregnancies), and maternal age on steroid concentrations were assessed.
  • The dataset was split into a Test group for model development and a Validation group to assess predictive performance.
  • Bivariate and multivariate statistical models were constructed to predict fetal sex based on steroid levels.

Key Findings

• Fetal sex emerged as the primary factor significantly affecting conjugated steroid profiles (p-values between 0.01 and 0.0001), confirming strong biological differences.
• Conjugated estrogens and DHEA-S patterns differed between female and male pregnancies:
  • Female pregnancies exhibited higher concentrations of classical conjugated steroids around block 5 (approximately mid-gestation).
  • Male pregnancies showed a delayed elevation in these steroids that persisted until term.
  • Equilin derivatives (a type of estrogen) were consistently lower in male pregnancies throughout gestation.
• Estrone sulfate (E1S) alone provided reasonable fetal-sex prediction accuracy:
  • Area Under the Curve (AUC) ranged from 0.710 to 0.865 between blocks 5 and 10, indicating fair to good performance.
• More robust multivariate models using the full panel of conjugated steroids markedly improved accuracy:
  • AUC values ranged from 0.747 to 0.912 across blocks 5-11, reflecting strong discriminatory power.
  • This approach outperformed univariate models and represents a viable alternative to ultrasound-based sexing in later pregnancy stages.
• Other factors such as breed, parity, and maternal age had less influence on steroid profiles compared to fetal sex, supporting the specificity of the method.

Implications and Applications

• The validated LC-MS/MS method enhances understanding of the endocrine interactions between fetus and placenta in horse pregnancy.
• This steroid profiling approach provides a non-invasive, objective, and reliable tool for fetal-sex determination from mid-gestation to term.
• Improved prediction accuracy could:
  • Assist breeders and veterinarians in planning and management decisions.
  • Reduce reliance on operator-dependent ultrasound methods, especially late in gestation when imaging is challenging.
  • Potentially be adapted to automated or routine laboratory settings for wide accessibility.
• The method may also set the stage for exploring other aspects of equine pregnancy health and fetal well-being through endocrine biomarkers.

Conclusion

• This study successfully demonstrated that maternal serum profiles of conjugated estrogens and androgens, analyzed via a rigorously validated LC-MS/MS method, can accurately and non-invasively determine equine fetal sex from mid-gestation to term.
• The multivariate steroid profiling approach offers improved accuracy compared to single analyte models and can complement or replace current fetal-sexing techniques.
• Overall, this work represents a valuable advancement in equine reproductive endocrinology and practical veterinary diagnostics.