Artificial insemination (AI) is a central tool for genetic improvement in contemporary cattle production, enabling the rapid dissemination of genetically high-merit germplasm through the globally standardized infrastructure underpinning frozen bull semen production and distribution (1). Despite advances in semen processing, cryopreservation, extender design, and quality control, pregnancy rates following AI still vary markedly across farms and seasons, and also among bulls and technicians (2–4). Epidemiological data further suggest that favourable laboratory-based semen quality metrics do not consistently translate into high field fertility, a pattern that may reflect the combined influence of environmental conditions and sire-specific biological differences that are incompletely captured by routine assays (5, 6).

Because dairy- and beef-oriented cattle populations have been subject to long term selection under differing production goals, they may also differ in reproductive phenotypes associated with sperm structure, metabolic traits, and resilience to cryoinjury (7, 8). Comparative studies suggest that production type can be reflected in post-thaw sperm quality profiles. In some cohorts, dairy bulls have tended to show higher plasma membrane integrity and a greater proportion of morphologically normal sperm, alongside higher indicators of mitochondrial respiratory activity and lower DNA fragmentation indices than beef bulls (8, 9). Complementing these univariate contrasts, multivariate approaches further indicate that bull type can contribute measurably to the overall semen quality signature beyond standardized processing conditions (10, 11). It is also important to note that production type often co-varies with breed composition and management context, which may jointly shape observed differences under field conditions.

These observations are compatible with the hypothesis that biology associated with breed or production type may modulate cryotolerance through differences in membrane composition and stability, mitochondrial function, and susceptibility to oxidative stress, factors repeatedly implicated in variation in sperm freezability among bulls (5, 12). Classical comparisons between beef-type and dairy-type bulls have also reported lower motility, viability, and normal morphology, together with a higher prevalence of specific primary sperm abnormalities in certain beef breeds, patterns that may suggest differences in spermatogenic efficiency and in sensitivity to physiological or environmental stressors (7, 9). Taken together, the available evidence supports the view that semen from dairy versus beef bulls may diverge not only in kinematic behaviour but also in functional attributes relevant to fertilizing potential, including membrane and acrosomal status, mitochondrial competence, and chromatin integrity, which may contribute to variability in field fertility even when semen is produced under standardized laboratory protocols (5, 6).

Computer-assisted sperm analysis (CASA) is widely used for objective semen evaluation, enabling quantitative assessment of sperm kinematics such as average path velocity (VAP), straight-line velocity (VSL), and curvilinear velocity (VCL), together with related descriptors including linearity (LIN), amplitude of lateral head displacement (ALH), and beat-cross frequency (BCF). Relative to conventional subjective microscopic assessment, CASA can improve measurement objectivity and within-laboratory repeatability when sampling conditions and instrument settings are carefully standardized, although comparability across systems and settings can still vary (13). Although some studies have reported significant associations between CASA derived kinematic parameters and fertility outcomes (3), others highlight the multifactorial determinants of reproductive success and suggest that individual kinematic variables, considered in isolation, offer limited predictive utility under commercial field conditions (6). This has motivated the use of multivariate analytical strategies and sperm subpopulation profiling as potentially more informative frameworks for linking motion phenotypes to fertility-relevant function.

Much of the existing literature has examined a single production type, often dairy cattle, or has evaluated semen primarily through in vitro assessments performed under controlled laboratory conditions. Consequently, it remains unclear whether production type associated differences in sperm kinematics and related functional attributes translate into meaningful variation in reproductive performance in commercial field settings. Field studies that jointly consider production type, detailed kinematic semen parameters, and pregnancy outcomes while maintaining comparable management, environmental exposure, and insemination procedures appear limited, despite their practical importance for AI program refinement. This evidence gap complicates the generalization of laboratory-derived semen quality indicators and reduces confidence in their use for fertility prediction across heterogeneous bull populations. It was hypothesized that frozen–thawed semen from dairy and beef bulls exhibits distinct post-thaw quality and kinematic profiles and that these differences are associated with post-AI pregnancy outcomes. The objective of this study was to compare post-thaw spermatological and CASA-derived kinematic traits of frozen semen from Holstein (dairy) and Angus (beef) bulls and to explore whether between-bull variation in these laboratory traits is associated with field pregnancy outcomes within the production systems in which the semen was used. As an observational study conducted under routine commercial conditions, this work aims to explore these relationships rather than establish causal attribution between sire production type and pregnancy success. This study aims to exploratorily evaluate the associations between spermatological parameters measured under commercial field conditions and fertility outcomes.

In the present study, spermatological characteristics of frozen–thawed semen and field fertility outcomes were evaluated in dairy and beef bulls under commercial insemination conditions. The most marked between-group difference was observed in motility, with semen from beef bulls showing substantially higher total motility and higher progressive motility than semen from dairy bulls. In contrast, the kinematic profile did not indicate a consistent advantage of one production type across velocity descriptors, and several velocity measures showed limited separation between groups. This pattern is compatible with previous reports suggesting that individual CASA-derived kinematic variables often have restricted predictive value for field fertility when considered in isolation, particularly in multifactorial commercial settings (3, 14).

Interpretation of semen type effects on fertility in this dataset needs to be made in the context of the female populations in which inseminations occurred. Semen from dairy bulls was used in high-yield dairy females, whereas semen from beef bulls was used in beef females, and the two female groups differed markedly in daily milk production (18.47 ± 0.38 vs. 6.21 ± 0.19 kg/d). Given the well-described association between high milk yield and reduced reproductive efficiency in dairy cattle, the lower pregnancy performance observed in the dairy female group could plausibly reflect female-side physiological constraints, including a higher likelihood of negative energy balance, compromised follicular dynamics, and altered uterine conditions that may reduce early embryo survival, even when semen meets acceptable laboratory quality thresholds (15–17). Under this structure, differences in pregnancy outcomes should not be interpreted as evidence that CASA kinematics are biologically irrelevant; rather, they underscore that field fertility is jointly determined by semen characteristics and the metabolic and management context of the female, and that the latter may dominate the observable outcome in high-producing systems.

This framing is consistent with large-scale evidence for substantial between-bull variation in field fertility under broadly comparable management conditions (18), while acknowledging that the magnitude of any bull effect can be attenuated when conception is primarily constrained by female-related factors linked to production intensity. In practical terms, the coupling of semen type with female production type in the present field setting represents a potential source of confounding and should be considered when attributing differences in pregnancy outcomes to bull production type alone. Accordingly, these comparisons are observational and do not support causal attribution of pregnancy differences to bull production type.

A further observation concerned the dispersion of selected kinematic and spermatological measures across bulls within each production type. In this dataset, variability appeared more pronounced for some traits in dairy bulls, most notably for progressive motility, whereas other parameters showed comparable dispersion between groups. Any interpretation of between-bull variability should therefore remain trait-specific rather than assuming a uniform pattern across all CASA outputs. Differences in within-group dispersion could plausibly relate to population structure and breeding history, including differences in effective population size, the intensity and direction of selection, and the extent to which reproductive traits have been directly or indirectly targeted in breeding programs. Dairy populations, particularly Holstein, have been shaped by multi-trait selection indices that integrate production, health, and efficiency traits, while beef populations are often managed under different breeding objectives and selection pipelines. Such contrasts may contribute to heterogeneity in sperm phenotypes, although the present data do not allow causal attribution. Prior CASA-based comparisons across breeds and production types provide context for observed differences in sperm motion characteristics (19, 20), but the extent to which they explain within-type variability requires dedicated analyses.

To reduce cross-system heterogeneity, correlations between semen kinematics and fertility were examined within production type rather than in a pooled dataset. Pooling dairy and beef systems would combine semen measurements with substantially different female production contexts, which could introduce structural confounding and obscure kinematic-fertility relationships. The within-group approach therefore provides a more interpretable framework for relating variation in kinematic parameters to pregnancy rate within a given production setting.

Within this within-group framework, motility showed the most consistent positive association with fertility in both production types, although the strength differed. In beef bulls (n = 5; 190 inseminations with valid pregnancy outcome), pregnancy rate correlated positively with motility (r = 0.48). In dairy bulls (n = 6; 219 inseminations with valid pregnancy outcome), the corresponding association was stronger (r = 0.94). Beyond motility, several kinematic descriptors displayed marked direction changes between production types. Linearity (LIN) and wobble (WOB) were positively correlated with fertility in beef bulls (LIN r = 0.79; WOB r = 0.58) but negatively correlated in dairy bulls (LIN r = −0.59; WOB r = −0.79). Conversely, ALH and BCF showed negative correlations with fertility in beef bulls (ALH r = −0.59; BCF r = −0.70) and positive correlations in dairy bulls (ALH r = 0.65; BCF r = 0.48). Sperm concentration followed the same pattern of sign reversal, with a negative association in beef bulls (r = −0.49) and a stronger positive association in dairy bulls (r = 0.73). These within-type patterns indicate that while motility aligned with fertility in both contexts, several ratio-derived trajectory and beat-frequency descriptors were associated with fertility in production-type-specific directions.

The production-type-specific sign reversals observed for LIN, WOB, ALH, BCF, and concentration suggest that the fertility relevance of individual CASA descriptors may be context dependent rather than universal. Several of these indices are mathematically coupled or ratio derived, and they can reflect different underlying motion phenotypes depending on the distribution of velocity components, tracking settings, and the biological state of the sperm population at assessment. Within-group correlations were estimated on a small number of bulls, making coefficients sensitive to influential observations and limiting the strength of inference. For these reasons, the within-type association structure is best interpreted as exploratory evidence that the mapping between kinematic phenotypes and field fertility may differ between production contexts, rather than as definitive proof of opposite causal mechanisms.

Velocity descriptors are often assumed to be beneficial proxies for sperm performance, yet field data do not consistently support this assumption. In the present study, velocity-related CASA measures did not provide a stable or production-type-robust fertility signal, and interpretation of these descriptors therefore warrants caution. One mechanistic explanation that remains plausible, although it was not directly evaluated here, is that elevated velocity shortly after thawing can coincide with cryo-induced signalling changes that reduce functional longevity. Premature capacitation, often termed cryo-capacitation, has been described in frozen–thawed bovine semen and involves membrane and signalling alterations that resemble physiological capacitation (21). Under this framework, a subpopulation may display hyperactivation-like movement soon after thawing, which can appear favourable in CASA snapshots but may be coupled to accelerated energy expenditure and a shorter functional lifespan within the female reproductive tract. Hyperactivation-like motility has also been linked to increased metabolic demand and reactive oxygen species generation, which may exacerbate oxidative stress and compromise membrane integrity and chromatin stability (22). These mechanisms provide biological context for why velocity-related CASA metrics may not map straightforwardly onto field pregnancy outcomes.

More broadly, these results indicate that production type related differences in semen characteristics are not necessarily mirrored by proportional differences in field fertility. Conception in commercial systems reflects a multifactorial process in which semen attributes interact with female physiological status, estrus detection and insemination timing, environmental conditions, and herd level management. Accordingly, favourable laboratory-derived semen metrics, even when they capture clear contrasts between bull groups, may be insufficient on their own to ensure high conception rates under field conditions.

Several limitations should be acknowledged. First, although a total of 413 inseminations were performed, pregnancy outcome was available for 409 inseminations, and the number of bulls contributing semen to the study was limited (n = 11). This constrains inference at the bull level and warrants confirmation in larger and more diverse bull populations, particularly for within-type association analyses. Second, inseminations were carried out under typical farm conditions, with semen from dairy bulls used in dairy females and semen from beef bulls used in beef females. As a result, female production type and physiological status could not be experimentally standardized, and residual confounding by female-related factors cannot be excluded. At the same time, this structure reflects routine breeding practice and increases ecological validity. Differences between female groups were quantified using milk yield records and were considered when interpreting fertility outcomes in relation to semen characteristics. Because the production-type contrast also corresponds to a breed contrast in the present dataset, generalisation to beef and dairy bulls more broadly should be made cautiously. Given the field-based design, the study is subject to structural confounding because semen type was not assigned independently of the female production context. Beef-bull semen was used in beef females and dairy-bull semen in high-yielding dairy females, coupling semen type with female physiology and herd-level management conditions. Consequently, pregnancy-rate differences cannot be experimentally attributed to semen characteristics alone, and the findings should be interpreted as context-dependent associations under commercial conditions rather than causal effects of bull production type.

Future studies in larger bull cohorts could apply hierarchical modelling frameworks to quantify the independent contributions of semen traits and female-side factors to pregnancy outcome.

Taken together, the present findings suggest that the association structure between CASA-derived kinematic descriptors and fertility can differ by production context, while motility remains a consistently informative correlate within both beef and dairy bulls. Practical evaluation of frozen semen is therefore likely to benefit from multidimensional assessment strategies that integrate kinematic profiles with broader functional indicators and that explicitly consider the physiological and management context of the inseminated females.