An imbalance between reactive oxygen species (ROS) and antioxidants, known as oxidative stress, causes damage to proteins, lipids, and nucleic acids. As such, oxidative stress contributes to the development of cancer and numerous diseases, including cardiovascular, metabolic and neurodegenerative disorders. Bovine milk is one of the most valuable food sources of antioxidants, which are found in all fractions: aqueous (e.g. vitamin C, microelements: Sn, Zn, Fe, Mn), lipid (e.g. vitamins E, A, D, β-carotene) and protein (e.g. β-lactoglobulin, lactoferrin) (Stobiecka et al., Reference Stobiecka, Król and Brodziak2022). Antioxidants present in milk are both biologically and technologically significant because they help preserve the chemical composition and physical properties of milk during processing and cheese ripening, extending the shelf life of dairy products. Moreover, they prevent the formation of undesirable flavours by protecting lipids from autoxidation (Van Aardt et al., Reference Van Aardt, Duncan, Marcy, Long, O'Keefe and Nielsen-Sims2005). Milk antioxidants could be classified as enzymatic [superoxide dismutase (SOD), glutathione peroxidase (GSHPx), catalase (CAT)] and non-enzymatic compounds (casein, whey protein, vitamins, minerals) (Elshaghabee et al., Reference Elshaghabee, AA, Ambrósio, de Gouveia and Singh2023). While it is possible to determine the activity of selected antioxidants individually, the complexity of the milk antioxidant system results in total antioxidant capacity (TAC) assays being the preferred approach. The TAC of milk and dairy products can be evaluated by one of the three well known methods: the 2,2-diphenyl-1-picrylhydrazyl radical scavenging assay (DPPH), the Trolox-equivalent antioxidant capacity assay based on ABTS•⁺ radical cation decolourisation (ABTS/TEAC), and the ferric-reducing antioxidant power assay (FRAP). The ABTS/TEAC method is widely used because of its simplicity, speed and sensitivity (Cloetens et al., Reference Cloetens, Panee and Akesson2013; Mishra et al., Reference Mishra, Shah, Mokashe, Chavan, Yadav and Prajapati2015). Additionally, this method allows the assessment of the activity of both enzymatic and nonenzymatic antioxidants in raw, whole milk.

The antioxidant capacity of cow's milk is influenced by multiple factors, among which the animals’ diet, housing conditions and genetic factors play a key role. The results of some studies suggest the existence of breed-related differences (Mode et al., Reference Mode, Magaji, Nzoniwu, Wamakko, Ahmad and Dandare2023; Annie et al., Reference Annie, Karthiayini and Lucy2025) as well as the effect of individual genes and age in determining this trait (Pokorska et al., Reference Pokorska, Poskart, Kułaj, Ochrem, Dusza, Gil, Świętek and Makulska2017). Gene-encoding proteins crucial to the immune response, such as osteopontin (OPN), chemokine receptor type 1 (CXCR1) and L-selectin (SELL), may modify the content of bioactive compounds in milk through their involvement in inflammatory processes.

Osteopontin is involved in regulating inflammatory cell migration and activity. The single-nucleotide substitution in the OPN gene (c.495C > T; rs109659827) causes an amino acid change in the encoded protein (Pro→Leu). In previous work (Bauer et al., Reference Bauer, Kułaj, Sawicki and Pokorska2023), this polymorphism was reported to be significantly associated with resistance to subclinical ketosis in dairy cows, which is closely related to immunity and oxidative stress (Karimi et al., Reference Karimi, Mohri, Seifi, Azizzadeh and Heidarpour2015). Zhang et al. (Reference Zhang, Heiribaut, Jing, Stefańska, Vandaele, De Neve and Fievez2023) showed that cows with ketosis had lower values of antioxidant indices and higher levels of oxidative stress markers compared to healthy cows.

Chemokine receptor type 1 plays a role in neutrophil chemotaxis. The rs1052336006 (c. + 365 T > C) in the bovine CXCR1 gene is a missense change leading to an amino acid substitution (Val→Ala) in the protein sequence. Studies by other authors linked this polymorphism to somatic cell count in milk (Zhou et al., Reference Zhou, Wang, Ju, Zhang, Huang, Qi, Hou, An, Zhong and Wang2013) and to the incidence of mastitis in dairy cattle (Pokorska et al., Reference Pokorska, Dusza, Kułaj, Żukowski and Makulska2016b). Furthermore, in silico modelling (Guzzi et al., Reference Guzzi, Oliveira, Amaro, Tavares-Filho and Gabriel2020) showed that this amino acid substitution induces local conformational changes in CXCR1 protein, potentially influencing the post-translational folding mechanism and its biological activity. The CXCR1 rs1052336005 (c.291C > T), in turn, is a synonymous variant. Although no direct associations between c.291C > T and udder health traits have been reported, haplotype-based analyses (Zhou et al., Reference Zhou, Wang, Ju, Zhang, Huang, Qi, Hou, An, Zhong and Wang2013) and multilocus genotype combinations (Muslimova et al., Reference Muslimova, Abdualiyeva, Shaugimbayeva, Orynkhanov and Ussenbekov2024) including this SNP associated to somatic cell count and susceptibility to subclinical mastitis in cattle.

L-selectin plays a crucial role in leukocyte adhesion. The SELL gene polymorphism (rs41803917; c.567 T > C), although a synonymous variant, in studies by other authors has been reported to be associated with somatic cell count in cows' milk (Dusza et al., Reference Dusza, Pokorska, Makulska, Kułaj and Cupiał2018) and with resistance to subclinical mastitis (Muslimova et al., Reference Muslimova, Abdualiyeva, Shaugimbayeva, Orynkhanov and Ussenbekov2024).

Due to the close association of these genes with immune mechanisms in dairy cows, their polymorphisms may not only determine disease susceptibility/resistance but also indirectly influence the oxidative status of milk. This preliminary study focused on the polymorphism analysis of four SNPs of known location and potential functional relevance: OPN (c.495C > T; rs109659827), CXCR1 (c. + 291C > T; rs1052336005), CXCR1 (c. + 365 T > C; rs1052336006) and SELL (c.567 T > C; rs41803917), and the total antioxidant capacity (TAC) of milk from Polish Holstein–Friesian black-and-white cows.

Material and methods

Sampling method

Multiparous Polish Holstein dairy cows, in their 7th to 10th lactations, characterised by an average daily milk yield of approximately 26 kg/cow/day in the preceding lactation (2023) were selected for the study. The research material comprised milk samples collected from 519 Polish Holstein–Friesian black-and-white cows, maintained in a free-stall system and fed a total mixed ration and milked three times daily. Animals were sourced from a large farm located in south-western Poland and were clinically healthy. The Afifarm management program (Afimilk®) was used for daily herd management and data storage, including AfiLab.

Laboratory analysis

Deoxyribonucleic acid (DNA) was isolated from milk samples (Pokorska et al., Reference Pokorska, Kułaj, Dusza, Żychlińska-Buczek and Makulska2016a) and used in PCR reactions to amplify selected gene segments (C1000 Thermal Cycler; Bio-Rad, CA, USA). Genotyping at the studied genomic loci employed a PCR-RFLP protocols, which involved primer sequences, restriction enzymes, reaction conditions, and thermal profiles based on previously published studies (Pokorska et al., Reference Pokorska, Dusza, Kułaj, Żukowski and Makulska2016b; Dusza et al., Reference Dusza, Pokorska, Makulska, Kułaj and Cupiał2018; Bauer et al., Reference Bauer, Kułaj, Sawicki and Pokorska2023) (Supplementary Tables S1 and S2). PCR-RFLP products were analysed in a 3.5% agarose gel (Merck, Germany) containing ethidium bromide (0.5 µg/mL). Electrophoresis (100 V, 90 min) was performed using a Mini-PROTEAN system (Bio-Rad) with the pUC19 DNA/MspI marker (Thermo Scientific, USA).

Milk samples were also used to measure TAC via the ABTS/TEAC method (Supplementary Methods). TAC values were determined directly in whole milk samples. The ABTS/TEAC assay is a spectrophotometric method which is based on the ability of antioxidants to neutralise the cation radical formed from ABTS •⁺ (2,2'-azino-bis (3-ethylbenzothiazoline-6-sulfonic)) under the influence of sodium persulfate, which is manifested by a decrease in the solution absorbance (Stobiecka et al., Reference Stobiecka, Król and Brodziak2022). The Hardy–Weinberg equilibrium calculator determined allele and genotype frequencies and Hardy–Weinberg equilibria.

Data analysis

The data were subjected to multivariate statistical analysis using Statistica®13 software (StatSoft®, Poland). All data had a normal distribution. To evaluate the association between selected genotypes (OPN, CXCR1, SELL) and the TAC of milk, analysis of variance (ANOVA) and Dunnett's post hoc test were employed, conducted at significance level of 0.05. Dunnett's test is a post hoc statistical test used after analysis of variance. It compares multiple treatment groups with a control and acts similarly to a t-test. The following formula gives a value that is used to compare mean differences:

\begin{equation*}t = \,\frac{{{{\bar X}_i} - \,{{\bar X}_c}}}{{SE}}\end{equation*}

Where:

${\bar X_i}$ = mean of treatment group i

${\bar X_c}$ = mean of the control group

n_i = sample size of treatment group i

n_c = sample size of the control group

SE = standard error of mean

Results

Analysis of individual loci showed that the herd was in genetic equilibrium, although the OPN locus was at the borderline of statistical significance (Table 1). The mean TAC of milk samples was 3.283 ± 1.616 mmol Trolox equivalents (TE)/L, ranging from 0.251 to 8.579 mmol TE/L. A considerable variability was observed, likely reflecting biological and genetic differences between cows. All values passed the quality control criteria (triplicate coefficient of variation ≤ 5%). No significant associations were found between individual genotypes and TAC values. Among the four loci analysed, OPN (c. 495C > T; rs109659827), CXCR1 (c.291C > T; rs1052336005), CXCR1 (c. + 365 T > C; rs1052336006) and SELL (c.567 T > C; rs41803917), a total of 55 genotype combinations were identified. Based on Dunnett's test, TAC depended on the combination of genotypes at the analysed loci, although variation at the OPN locus (c. 495C > T; rs109659827) did not substantially affect the trait under study. In the analysis of the remaining three loci, 21 genotype combinations were identified, with the TT (CXCR1- c.291C > T; rs1052336005) –TT (CXCR1- c. + 365 T > C; ss1052336006) –TT (SELL- c.567 T > C; rs41803917) combination significantly associated with a higher mean TAC (4.593 mmol TE/L) (P < 0.039). A higher value (5.198 mmol TE/L) was only recorded for the CT–CT–CT genotype combination, which was observed in a single individual (Supplementary Table S3).

Table 1.

Frequencies of alleles and genotypes at the OPN, SELL and CXCR1 loci

P-value indicates the result of the Hardy–Weinberg equilibrium (HWE) test;.

P -values < 0.05 suggest deviation from HWE.