Methanol was purchased from Thermo Fisher Scientific Co., Ltd. (81 Wyman Street, Waltham, MA, 02454), and anhydrous ethanol was purchased from Sinopharm Group Chemical Reagent Co., Ltd. (Shanghai, China).

The animal study was approved by the Ethics Committee of the College of Animal Science and Technology, China Agricultural University (AW51704202-1-1). The animal study was conducted at Sanshi Dairy Farm in Changping District, Beijing, China.

The milk samples were obtained from each cow at 5:00 on the sampling day and each sample was divided into two tubes. One tube was mixed with potassium dichromate and stored at 4 °C for Dairy Herd Improvement (DHI) test. The other tube was stored at −20 °C for testing the MT content.

The DHI test was conducted at the Dairy Herd Improvement Testing Station of the Beijing Livestock Station. Fifty microliter of potassium dichromate mixed milk was used for the DHI test. The fat, protein, lactose, and milk UN content in the milk were measured using a MilkoScan FT+ automatic analyzer (Denmark Serial No.9175049, Part No.60027086), and the SCC in the milk was measured using a FossomaticTM FC automatic analyzer (Denmark Serial No. 975377, Part No. 60002326).

The milk samples were ultrasonicated for 30 min at 37 °C, then, 1 ml of milk sample was added to 4 ml of methanol, vortexed for 5 min, and then centrifuged 12,000 rpm for 10 min at 4 °C. The supernatant was filtered through a 0.22 μm filter into a brown vial. MT was measured with Agilent 6470 liquid chromatography-tandem MS (LC-MS-MS; Agilent Technologies, Santa Clara, CA, USA).

On the sampling day, before morning feeding (5:30 am), whole blood was collected from the tail vein of cows and transferred into coagulation-promoting tubes. The samples were centrifuged at 3,000 rpm for 10 min at 4 °C, and the serum was then transferred into 5 ml centrifuge tubes and stored at −20 °C.

Five hundred microliter of serum was mixed with 2 ml of methanol, followed by vigorous shaking in a vortex oscillator for 10 min and then it was allowed to stand at −20 °C for 30 min. Afterward, the supernatant was extracted using a low-temperature high-speed centrifuge 12,000 rpm for 10 min at 4 °C, and filtered through a 0.22 μm filter into a brown vial. MT was measured with Agilent 6470 liquid chromatography-tandem MS (LC-MS-MS; Agilent Technologies, Santa Clara, CA, USA).

The SOD, MDA, and T-AOC were tested using biochemical methods. LH, E2, and P4 were tested using enzyme-linked immunosorbent assay (ELISA). PRL, COR, IL-6, IL-10, and TNF-α were also tested using ELISA. All the testing reagents were procured from Beijing Solarbio Science & Technology Co., Ltd and operated according to the instructions manual.

The data were presented as mean ± SEM. If not specifically stated, the difference between two groups was analyzed using t-tests, while differences among three or more groups were analyzed using one-way ANOVA followed by the Duncan's multiple range test. Pearson's correlation analysis was employed to determine the degree of association between the two variables. Statistical analysis was performed using GraphPad Prism 8.0.2 software. A value of P < 0.05 was considered statistically significant. The confidence interval is 95%.

To investigate the effects of different light wavelengths on milk yield and quality, we divided the cows into three groups exposed to darkness, yellow light, or red light at night. The results showed that the MT content in the milk of the cows under yellow light management was significantly higher than in those in the other two groups, while no significant difference was observed between the red light treated cows and the control cows (Figure 1A). There were no significant differences in milk fat percentage and milk UN levels among the groups (Figures 1B, C), but the milk protein percentage in the cows treated with yellow light was significantly higher and the SCC was significantly lower than those in the other two groups (Figures 1D, F). And the lactose percentage in the cows treated with yellow light was significantly higher than the control group (Figure 1E). However, these parameters had no significant differences between the red light and the control group (Figures 1D–F). In addition, the daily milk production of cows with the yellow light illumination was significantly higher than those in the other two groups (Figure 1G).

To confirm these observations, these parameters were compared in the same group before and after yellow light treatment. The results showed that the milk MT content was significantly higher after the yellow light exposure than that before the exposure (20.49–29.22 pg/ml; Figure 2A). The similar results was observed in the daily milk yield (32.39–37.58 kg; Figure 2B), the milk fat percentage (3.53%−3.70%; Figure 2C), the milk protein percentage (3.13%−3.41%; Figure 2D) and the lactose percentage (4.76%−5.11%; Figure 2E), While the milk UN content was significantly reduced after yellow light exposure compared to the values before the exposure (14.16–13.70 mg/dl; Figure 2F), and SCC in the milk were also reduced (138.60–123.00 × 10³/ml), but the reduction of SCC did not reach significant difference compared before the exposure (Figure 2G).

To explore the relationship between MT and milk quality, we conducted correlation analyses between these indicators and MT levels. The results from the linear correlation analysis showed a significantly positive correlation between milk MT content and milk protein percentage (R = 0.624, P < 0.001) and lactose percentage (R = 0.394, P < 0.001; Figures 3A, C). In contrast, a highly significant negative correlation was observed between milk MT content and milk SCC (R = −0.380, P < 0.001; Figure 3E). No significant correlation was found between milk MT content and milk fat percentage (R = −0.030, P > 0.05; Figure 3B) and milk UN (R = −0.044, P > 0.05; Figure 3D).

We then examined the effect of yellow light on serum MT levels and antioxidant capacity in dairy cows by evaluating relevant indicators. The results showed that the overall level of serum MT in cows after yellow light exposure was significantly higher than that before their exposure (36.30–59.48 pg/ml; Figure 4A), but the yellow light exposure had little effects on the serum COR content compared to the level before exposure (Figure 4B). The yellow light exposure significantly elevated the serum SOD (111.80–117.60 U/ml; Figure 4C) and T-AOC (8.28–8.76 U/ml; Figure 4E) levels, but significantly lowered the serum MDA level (4.69–4.20 nmol/ml; Figure 4D).

To assess changes in immune function, we measured representative indicators: TNF-α, IL-6, and IL-10. Correlation analyses were then performed between MT and each of TNF-α and IL-6. The results showed that yellow light exposure significantly lowered the serum TNF-α (181.10–174.90 pg/ml; Figure 5A) and IL-6 levels (117.30–113.90 pg/ml; Figure 5C) compared to the values before exposure. However, yellow light exposure significantly increased the serum IL-10 level compared to the value before exposure (31.18–32.86 pg/ml; Figure 5E). The results of the linear correlation analysis showed a significantly negative correlation between serum MT and TNF-α (R = −0.3510, P = 0.005; Figure 5B) as well as IL-6 levels (R = −0.5020, P < 0.001; Figure 5D).

To investigate the effects of yellow light treatment on reproductive hormones and performance, we analyzed the changes in E2, P4, LH, and PRL levels before and after the treatment. We also tracked the reproductive metrics of cows exposed to yellow light compared to those not exposed. The results showed that there were significant differences in the levels of LH (5.336–5.708 mIU/ml) and PRL (218.4–246.1 μIU/ml) after yellow light exposure (Figures 6C, D). However, the yellow light exposure significantly decreased the P4 level compared to the value before the exposure (Figure 6B). There was no significant difference in the E2 level (Figure 6A). Compared to the control group, cows exposed to yellow light showed significant improvements in reproductive metrics (Table 1): the interval to first postpartum estrous cycle was shortened (61.72 ± 1.27 to 56.91 ± 1.14 days), the pregnancy rate after first-insemination was increased (23.68 ± 4.42% to 38.15 ± 5.00%), and the pregnancy days after first-insemination were reduced (96.84 ± 4.88 to 82.95 ± 4.50 days).

As early as 2006, Gnann proposed a method for producing high-MT milk, which involved initially using blue-enriched light to inhibit MT secretion, followed by exposure to at least one light source above 500 nm (capable of emitting red, orange, yellow, or mixed light) for 2 h to maximally reverse MT suppression, thereby yielding dairy products with elevated MT levels. Our findings are consistent with this approach, demonstrating that yellow light positively influences MT concentrations in milk. Moreover, we observed that yellow light was more effective than red light in enhancing MT levels in both milk and serum. These results are consistent with another study, which reported that yellow LED light increased serum MT levels in dairy cows compared to blue LED light (9). The effects rely on the melatonin synthesis mechanism in the pineal gland. In mammals, light information is detected by intrinsically photosensitive retinal ganglion cells (ipRGCs). This signal is transmitted via the retinohypothalamic tract (RHT) to the suprachiasmatic nucleus (SCN), which houses the circadian pacemaker (17). Through a polysynaptic neural pathway, the SCN controls the release of norepinephrine from sympathetic nerve endings in the pineal gland. This results in a rapid increase in the activity and expression of the melatonin-synthesizing enzyme AANAT, thereby promoting melatonin synthesis and secretion (18, 19).

Photoreceptors exhibit varying sensitivities to different light wavelengths. Specifically, melanopsin-containing ipRGCs function as non-visual photoreceptors (20). They show highest sensitivity to short-wavelength light (e.g., blue light at 480 nm) and lowest sensitivity to red light (18). Short-wavelength light signals (indicating daytime) received by melanopsin travel along the RHT to the SCN. The SCN releases GABA, which disinhibits the paraventricular nucleus (PVN) of the hypothalamus. This action suppresses the sympathetic pathway to the pineal gland, ultimately inhibiting melatonin synthesis. At night, the SCN ceases GABA production, lifting the inhibition on the downstream pathway (19). In our results, melatonin levels in the yellow light treatment group were markedly higher than in the dark control group. This suggests that yellow light may affect the pathway through multiple mechanisms. Firstly, it could suppress GABA synthesis, removing the inhibition on melatonin production. Secondly, it might enhance norepinephrine release from pineal sympathetic nerves, further stimulating melatonin synthesis.

Subsequent analysis demonstrated that yellow light treatment significantly enhanced milk yield, milk fat percentage, milk protein percentage, and lactose content, whereas milk UN decreased significantly and SCC exhibited a decreasing trend. Further correlation analysis indicated that changes in milk protein percentage, lactose content, and SCC were significantly associated with MT levels. Numerous studies support the beneficial role of MT in enhancing milk quality. For instance, Li et al. (21) reported that dietary supplementation with 120 mg of MT for 30 days increased milk fat and protein percentages while reducing SCC. Similarly, Wu et al. (22) demonstrated that exogenous MT administration improved Dairy Herd Improvement (DHI) parameters. Milk UN, as another important component in milk, reflects the balance of dietary nutrition and is commonly used to evaluate protein utilization efficiency. High milk UN levels indicate excessive dietary crude protein or insufficient non-structural carbohydrates in the diet, while low milk UN suggests a need to reassess sources of dietary protein and carbohydrates. Multiple studies have shown that dairy cow lameness (23), ketosis (24), and abnormal reproductive performance (25) are associated with milk UN levels. During our trial, although the structure of the dairy cows' diet remained unchanged, the decrease in milk UN and increase in milk protein indicate that yellow light improved the efficiency of protein utilization, which positively influenced the health status and reproductive performance of the cows.

In the present study, we also observed an increase in serum prolactin levels following yellow light treatment, which may directly contribute to the elevated milk yield, given the established lactogenic role of prolactin in dairy cattle (26). The secretion of prolactin may be simultaneously regulated by light and melatonin. Melatonin stimulates PRL expression and secretion by promoting the Nrg1/ErbB4 signaling pathway (27). However, previous research reported that exogenous melatonin inhibits prolactin secretion in mammary epithelial cells of dairy goats (28), and we speculate that this opposite effect is dose-dependent. Similar to melatonin, prolactin exhibits a circadian rhythm, increasing in darkness and decreasing under light exposure (29). In this study, we propose that yellow light directly elevates serum melatonin and prolactin levels, but an interaction between melatonin and prolactin is likely. In summary, using yellow light at night can both increase milk yield and improve milk composition, thereby comprehensively enhancing dairy cow milk performance.

Oxidative stress and inflammation influence each other (30). Oxidative phenomena can transmit and manage signals during the initial stages of inflammation (31). Meanwhile, the secretion of substantial reactive substances by inflammatory cells contributes to increased oxidative stress at the site of inflammation.

In this study, with the accumulation of days under yellow light treatment, the level of IL-10 in dairy cow serum gradually increased, while the levels of TNF-α and IL-6 progressively decreased. In addition, the MDA level was lower than the pre-treatment level, and the levels of SOD and T-AOC were elevated. These findings suggest that both immune and antioxidant capacities were improved after yellow light treatment. Research shows that the recovery rate of cow mastitis treated with antibiotics combined with low-intensity laser irradiation at 870–970 nm was significantly higher than that of the antibiotic-only treatment group, indicating that light can influence the body's inflammatory response (32). Patients with melasma treated with topical depigmenting serum combined with 577 nm yellow light laser had significantly higher serum T-AOC and SOD1 levels than those treated with topical depigmenting serum alone (33). These are consistent with our conclusion that yellow light can enhance the body's antioxidant and anti-inflammatory capacity. Subsequently, we found that serum MT was significantly negatively correlated with TNF-α and IL-6. MT is an effective antioxidant and anti-inflammatory agent (34). Therefore, we speculate that these changes induced by yellow light are closely related to increased MT levels in the body. Yu et al. (35) confirmed that MT exerts both antioxidant and anti-inflammatory effects on bovine mammary epithelial cells: it reduces TNF-α and IL-6 mRNA expression in LPS-stimulated cells via the TLR4 signaling pathway, thereby decreasing inflammatory responses; it also activates the Nrf2 antioxidant defense pathway to inhibit oxidative stress. Li et al. (21) fed mid-to-late lactation cows 120 mg of MT for 30 days and found significantly reduced serum TNF-α, IL-6 and MDA levels, while improved serum IL-10 and SOD levels. In this study, yellow light treatment increased melatonin levels, and melatonin exerts antioxidant effects through multiple pathways. On one hand, melatonin and its metabolites directly bind to free radicals, forming a cascade reaction that efficiently scavenges reactive oxygen species. On the other hand, melatonin activates the antioxidant defense system SOD, ultimately increasing T-AOC and decreasing MDA. In previous studies, melatonin has been shown to mediate anti-inflammatory effects via the nuclear factor kappaB pathway (36), TLR4 pathway (35), and Nrf2 pathway (34). Here, we only detected changes in inflammatory factors, and more specific mechanisms require further investigation.

In this study, yellow light treatment resulted in decreased serum P4 levels and increased PRL and LH levels in dairy cows, suggesting that the cows were in the transition phase from the luteal phase to the follicular phase. Combined with the shortened the interval to first postpartum estrous cycle, increased pregnancy rate, and reduced pregnancy days, we speculate that yellow light treatment enabled cows to recover more rapidly from the pregnancy-lactation physiological state, initiate normal ovarian cyclicity, and successfully complete subsequent pregnancy. We propose that these improvements in reproductive indices depend on the enhanced overall health status of cows after yellow light treatment, including improved protein utilization, immune function, and antioxidant capacity. This is because studies have shown that inflammation, oxidative stress, and malnutrition can adversely affect reproductive performance in dairy cows. For example, mastitis affects not only the udder but also extends to the entire reproductive system, leading to decreased conception rates, abnormal estrous cycles, early embryonic death or abortion, as well as increased days open and service index (number of artificial insemination (AI) required per conception) and reduced conception rates (37). Kumar et al. (37) speculated that this might be related to factors such as mastitis pathogens, infection duration, and excessive cytokine production during the disease process. Japanese Black cattle with poor reproductive performance exhibited significantly higher oxidative stress indices in uterine fluid, while the concentration of antioxidants in the uterus positively influenced the uterine environment and subsequent establishment of successful pregnancy in Japanese Black cattle (38). During pregnancy in dairy cows, the antioxidant status of uterine luminal cells, peripheral blood MT concentration, and local expression of antioxidant cell markers all influence embryo quality and pregnancy success (39). Studies have shown that both excessively high and low MUN concentrations are detrimental to dairy cow reproduction, with varying thresholds among breeds (40).