Rumen-protected methionine supplementation improves lactation performance and alleviates inflammation during a subclinical mastitis challenge in lactating dairy cows

This study aimed to evaluate the effects of rumen-pro-tected Met on lactation performance, inflammation and immune response, and liver glutathione of lactating dairy cows during a subclinical mastitis challenge (SMC). Thirty-two Holstein cows (145 ± 51 DIM) were enrolled in a randomized complete block design. At −21 d relative to the SMC, cows were assigned to dietary treatments, and data were collected before and during the SMC. Cows were blocked according to parity, DIM, and milk yield and received a basal diet (17.4% CP; Lys 7.01% MP and Met 2.14% MP) plus 100 g/d of ground corn (CON; n = 16) or a basal diet plus 100 g/d of ground corn and rumen-protected Met (SM, Smartamine M at 0.09% of dietary DM; n = 16), fed as a top-dress. At 0 d, the mammary gland's rear right quarter was infused with 100,000 cfu of Streptococcus uberis (O140J). Milk yield was re - corded twice daily from 0 until 3 d relative to SMC. Milk samples were collected during each milking from 0 to 3 d relative to SMC, blood samples were collected at 0, 6, 12, 24, 48, and 72 h relative to SMC. The mTOR pathway activation was assessed in immune cells in blood and milk samples by measuring quantity and phosphorylation status of mTOR-related proteins, including AKT, S6RP, and 4EBP1. For the ratio of phosphorylated to total AKT, S6RP, and 4EBP1, blood samples were collected at 0, 12, and 24 h, and milk samples at 24 h relative to SMC. Liver biopsies were performed at −10 d and 24 h relative to SMC for measurement of glutathione. Linear mixed models with repeated measures were used to analyze the results. There was a trend for greater milk yield per milking (+ 0.8 kg) and per day (+1.7 kg) after SMC in SM cows compared with CON. The DMI was not affected by dietary treatments. Reactive oxygen metabolites (ROM) were lower in SM cows than in CON. Milk somatic cell linear score was not affected by dietary treatments, and a score >4 at 24 h confirmed subclinical mastitis. The SM cows had greater milk fat percentage at 24 and 36 h post SMC, resulting in overall greater milk fat. Milk protein tended to be greater in SM cows than in CON. We ob - served greater liver glutathione in SM cows than in CON. Among inflammation biomarkers, ceruloplasmin was lower for SM cows compared with CON. In milk, greater pAKT: AKT and pS6RP: S6RP ratios were observed in immune cell populations from SM cows compared with CON. Blood neutrophils had a greater p4EBP1: 4EBP1 ratio in SM cows compared with CON. Overall, our results show that Met supplementation during an SMC positively affected milk performance, lowered the risk of oxidative stress, and attenuated inflammation partially by increasing liver glutathione and immune cells’ protein synthesis via mTOR signaling.


INTRODUCTION
Mastitis is a well-known disease that affects the dairy industry and results in significant financial losses, increased antibiotic usage, and reduced animal welfare (Carlén et al., 2006).Mastitis is mostly caused by pathogenic bacteria invading and multiplying in mammary gland tissues (Ruegg, 2017).The great impact of mastitis on the dairy industry has been mainly reflected in economic losses.Jones, (2009) estimated that the cost of Rumen-protected methionine supplementation improves lactation performance and alleviates inflammation during a subclinical mastitis challenge in lactating dairy cows mastitis in the US dairy industry is ca.$2 billion annually or 11% of the total value of US milk production, with an average of ca.$170/cow annually.Economic losses are mainly caused by the decrease in milk production and the increase in the use of veterinary care.
Subclinical mastitis is characterized by an elevated somatic cell count (SCC) in milk, despite the absence of clinical signs (Yang and Li, 2015, Adkins andMiddleton, 2018).Quantifying the losses of subclinical mastitis is challenging; however, it is widely acknowledged as the most economically detrimental form of mastitis for the dairy industry (Dahl et al., 2018;Romero et al., 2018;Cheng and Han, 2020).Additionally, subclinical mastitis is considered a reservoir of microorganisms that leads to the infection of other animals within the herd (Ruegg, 2012).
The immune response efficiency in dairy cattle is closely linked to their nutritional health.Nutritional needs vary throughout the production cycle, and any dietary deficiency can negatively affect their immune resilience.While the relationship between nutrient metabolism and immunity is intricate and not fully understood, it is widely acknowledged that an adequate supply of dietary micronutrients (e.g., vitamins and trace minerals) is crucial for optimizing immune responses and enhancing disease resistance (Sordillo, 2018).Similar to vitamins and minerals, essential amino acids such as Met positively impact the immune system in dairy cows, particularly in processes related to inflammation site recognition via L-selectin (Li et al., 2016), phagocytosis (Osorio et al., 2013), and oxidative burst (Zhou et al., 2016a).
The mammalian target of rapamycin (mTOR) pathway is the primary intracellular mechanism that detects nutrient levels.It plays a crucial role in protein synthesis as well as integrating metabolic signals during immune cell activation across various mammalian species (Weichhart et al., 2008).The baseline AKT/mTOR signaling pathway activation in bovine immune cells was found to decrease postpartum (Mann et al., 2018).However, mTOR activation seems to mediate the inflammatory response in immune cells through mTOR-target proteins, including eIF4E-binding protein 1 (4EBP1) and S6 ribosomal protein (S6RP) (Weichhart et al., 2008).Essential AA have been demonstrated to influence mTOR phosphorylation and, consequently, protein synthesis in lactating dairy cows (Appuhamy et al., 2011).For instance, Met has been commonly associated with increased protein synthesis in in vitro (Nan et al., 2014) and in vivo models (Zhou et al., 2016b;Pate et al., 2020).Methionine cannot only influence mTOR through phosphorylation (Nan et al., 2014) but also through alternative pathways such as ubiquitination (Xie et al., 2024) and gene transcription upregulation (Qi et al., 2024).In bovine immune cells isolated around calving, mTOR activation plays a crucial role in mounting an adequate immune response to LPS, which could be partially attributed to an optimal provision of EAA, glucose, and insulin (Sipka et al., 2020).
Glutathione is commonly recognized as a potent antioxidant that can alleviate oxidative stress by serving as a reducing agent for the glutathione peroxidase defense mechanism (Lu, 2009).Methionine is an upstream precursor of glutathione (GSH).Feeding rumen-protected Met (RPM) during the transition period has been consistently reported to increase GSH concentration in the liver of dairy cows (Osorio et al., 2014b;Zhou et al., 2016a;Batistel et al., 2018).Oxidative stress, along with altered nutrient metabolism and dysfunctional inflammatory response, form the basis of metabolic stress that is characterized as the underlying factor for dairy cow diseases, including mastitis (Sordillo and Mavangira, 2014).Therefore, it is plausible that supplementing RPM during a subclinical mastitis challenge will allow cows to face such inflammatory conditions with better metabolic health.
Although, previous studies have reported the effects of supplementing RPM during the transition period, RPM effects during a subclinical mastitis model in lactating dairy cows have not been evaluated yet.Therefore, the objective of this study was to evaluate the effects of RPM on lactation performance, immune response, and metabolism of lactating dairy cows during a subclinical mastitis challenge (SMC).

MATERIALS AND METHODS
All the procedures for this study were conducted in accordance with the protocol approved by the Institutional Animal Care and Use Committee at the South Dakota State University (Protocol no. 2002-014A).This study comprised of an initial mastitis test, a titration experiment, and a final subclinical mastitis experiment.

Initial subclinical mastitis test
A pilot experiment was conducted to determine if 5,000 cfu could induce subclinical mastitis (Moyes et al., 2009) under SDSU dairy farm conditions.The rear right quarter of 8 cows (n = 4/treatment) was infused with either saline as control or 5,000 cfu using Streptococcus uberis strain O140J.After the infusion, milk samples were taken at 0, 12, 24, 36, 48, 60, and 72 h relative to infusion to check the variation in Somatic Cell Count (SCC).Infused cows with Strep.uberis did not develop subclinical mastitis (Figure S1), showing an SCC lower than 200,000 cells/ ml.Based on these results, we determined that a titration experiment was needed to establish an optimal concentration of Strep.uberis to induce subclinical mastitis.

Titration experiment
A titration experiment was conducted in which 6 cows were used to determine the appropriate concentration of the bacteria to induce subclinical mastitis.Cows were divided into 3 groups of 2 cows each.The mammary gland's rear right quarter from the 3 groups was infused with 100,000 cfu, 1,000,000 cfu, and 10,000,000 cfu, respectively, using Streptococcus uberis strain O140J.Milk samples were taken before the intramammary infusion to confirm the cows were bacteriologically negative before the challenge.After the infusion, milk samples were taken at 0, 12, 24, 36, 48, 60, and 72 h relative to infusion to check the variation in Somatic Cell Count (SCC) and determine that the 100,000 cfu infusion was the most appropriate infusion rate to cause temporary subclinical mastitis (Figure S2).The effect of Strep.uberis infusion level did not affect (P = 0.40) the overall Log10 SCC; however, the 100,000 cfu infusion level was able to temporarily pass the 5.3 log 10 SCC threshold (200,000 SCC/mL) at 36 h post-infusion and declined below said threshold afterward.All inoculum propagations and challenge material preparations of Streptococcus uberis O140J for this project were produced by Research Technology Innovation labs (RTI, LLC, Brookings SD 57006, USA).

Animal management
The experiment was conducted from May to June 2021 at the South Dakota State University Dairy Research and Training Facility (Brookings, SD).Hobo Pro series Temp probes (Onset Computer Corp., Pocasset, MA) were used to record the air temperature and relative humidity in the pen during the experimental period.Cows were housed in a ventilated enclosed barn with access to mattress-free stalls and fed using an individual gate system (American Calan, Northwood, NH, USA), and individual orts were collected once a day before feeding to determine feed intake.The feed offered was adjusted daily to achieve 5 to 10% refusal.The dry matter content of feed ingredients was determined once a week throughout the experiment, and diets were adjusted accordingly to maintain formulated DM ratios.
Experimental design and treatments.Thirty-two midlactation Holstein cows from 50 to 145 d in milk (145 ± 51 DIM; mean ± SD), including 28 multiparous and 4 primiparous cows, were used in a randomized complete block design.Cows were blocked according to parity, days in milk, and milk yield.Two cows (n = 1/trt) were removed from the trial due to abnormal eating behaviors.A total of 30 multiparous Holstein cows were fed experimental treatments from −21 to 7 d relative to SMC consisting of a basal diet (Table 1) plus 100g of ground corn (CON; n = 15) or basal diet plus Smartamine M (SM; n = 15) at a rate of 0.09%, and 100g of ground corn.Methionine supplement was top-dressed once a day within ~10 min after morning feeding from −21 to 7 d relative to SMC.The Smartamine M contains 75% DL-Met, physically protected by a pH-sensitive coating, which is considered to have a Met bioavailability of 80% (Schwab, 1995).Basal diet was formulated using the CNCPS model contained within the Agricultural Modeling and Training Systems (AMTS) CattlePro diet-balancing software (version 4.16.1,AMTS LLC, Lansing, NY, USA) to meet the requirements of the average cow in the group (Table 1).Before enrollment in the trial, cows were offered a typical diet for lactating cows without any RPM source.We evaluated the power of the study to detect differences in blood ceruloplasmin, an important acute phase protein secreted by the liver during an inflammatory response.We used variance data from a previous trial (Osorio et al., 2014b) and found that 32 (16/treatment) would provide 80% power to detect a 0.31 difference in blood ceruloplasmin.

Subclinical mastitis challenge and health indicators
Four days before SMC, foremilk samples from all quarters of each cow were cultured to verify that all experimental animals were bacteriologically negative.After 21 d on dietary treatments, immediately after the morning milking on d 22, the mammary gland's rear right quarter of all cows was infused with 100,000 cfu of Streptococcus uberis strain O140J in 5 mL of inoculum administered via a sterile disposable syringe fitted with a sterile teat cannula using the full insertion infusion method.Before inoculation, challenged teats were rigorously cleaned with cotton balls containing 70% isopropyl alcohol.Starting at 72 h post-inoculation, after all, samples had been collected, infected quarters were aseptically infused with 125 mg of ceftiofur HCl (SpectraMast LC, Pfizer Animal Health, Kalamazoo, MI) twice daily for 4 consecutive days.During the first 3 d of antibiotic treatment, cows were also administered 30 mL of procaine penicillin G (300,000 IU/mL, i.m.; US Vet, Hanford Pharmaceuticals, Syracuse, NY) once a day.The clinical response to SMC Strep.uberis challenge was monitored through rectal temperature, heart rate, and respiratory rate at 0,4,8,12,16,20,24,32,40,48,56,64, and 72h post-challenge.

Milk sample collection and analysis
Cows were milked twice daily, and the milk yield was recorded at each milking during the adaptation and experimental periods.Consecutive morning and evening milk samples were collected 1 d/wk during the experi- mental period.Composite milk samples were performed in proportion to milk yield at each milking, preserved with bronopol and natamycin (Broad Spectrum Microtabs II, Advanced Instruments), and analyzed for fat, protein, lactose, solids, MUN, and SCC using Fourier-transform infrared spectroscopy technology (DairyOne).

Feed sample
Total mixed ration samples were collected weekly and frozen at −20°C after DM analysis until further nutrient profile analysis.Monthly composites were analyzed for DM, CP, NDF, and ADF contents, while NEL was calculated using wet chemistry methods at a commercial laboratory (Dairy One; Ithaca, NY; https: / / dairyone .com/download/ forage -forage -lab -analytical -procedures/ ).

Blood collection and analysis
Blood samples were collected from the coccygeal vein using a 20-gauge vacutainer needle (Becton Dickinson, Franklin Lakes, NJ) at 0, 6, 12, 24, 48, and 72 h relative to SMC for blood biomarkers analysis.Additional blood samples were taken at 0, 12, and 24 h relative to SMC flow cytometry.Blood samples were collected into evacuated tubes (BD Vacutainer, Becton Dickinson, Franklin Lakes, NJ) containing either serum clot activator or lithium heparin.After collection, tubes that contained lithium heparin were placed on ice while tubes with serum clot activator were kept at 21°C until centrifugation.Serum and plasma were obtained by centrifugation at 1,300 × g for 15 min at 21°C and 4°C, respectively.The aliquots were frozen at −80°C until further analysis.
Albumin, cholesterol, and glucose were analyzed using the IL Test purchased from Instrumentation Laboratory Spa (Werfen Co., Milan, Italy) in the ILAB 600 clinical auto-analyzer (Instrumentation Laboratory, Lexington, MA), following the procedures described previously by Mezzetti et al. (2019).Ceruloplasmin was determined based on Sunderman and Nomoto (1970) with modifications described by Jacometo et al., (2015).Antioxidant potential was assessed as ferric-reducing antioxidant power (FRAP) using a colorimetric method (Benzie and Strain, 1996).Reactive oxygen metabolites (ROM) were analyzed with the d-ROMs-test (cod.MC002), purchased from Diacron (Grosseto, Italy).Paraoxonase was analyzed according to methods described by Trevisi et al. (2013).Myeloperoxidase was determined via colorimetry based on the reaction of MPO contained in the plasma sample with H 2 O 2 , which forms H 2 O and O − ; the O − dianisidine dihydrochloride, and electron donor, reacts with the O − , releasing H 2 O and a colored compound (Jacometo et al., 2015).Nonesterified fatty acids and BHB were measured using kits from Wako (Chemicals GmbH, Neuss, Germany) and Randox (Randox Laboratories Ltd., Crumlin, UK), respectively, following the procedures described previously by Mezzetti et al. (2019).

Flow cytometry analysis
Lysing, fixing, and permeabilizing.Milk and blood cells were isolated, lysed, fixed, and permeabilized using the BD Phosflow Lyse/Fix buffer I (5 × concentration) and BD Perm buffer III (Cat.No. 558049 and Cat, No. 558050 respectively, BD Biosciences, San Jose, CA) following protocols previously described by Sipka et al. (2020), with modifications.Briefly, 1.9 mL of whole blood and 1x10 7 milk cells (suspended in 1 × PBS 1% BSA) were lysed and fixed in 38 mL of 1:5 BD Lyse/Fix buffer I at 37°C for 10 min.Subsequently, samples were centrifuged for 8 min at 500 × g, and the supernatant was discarded.Samples were washed one more time with 38 mL of 1 × PBS and centrifuged at 10 min at 500 × g.The supernatant was again discarded, and cell pellets were then placed on ice.Samples were permeabilized by adding 1.5 mL ice-cold BD Perm buffer III and incubating for 30 min on ice.After incubation, cells were washed with 850 µL of 1 × PBS, centrifuged at 250 × g, and froze at −80°C until antibody labeling.
Antibody labeling.Aliquots containing 1 × 10 6 cells each were labeled with an antibody to detect monocytes and macrophages (APC anti-human CD14, BioLegend, Inc., San Diego, CA, USA), and neutrophils (anti-bovine CH138A IgM, Monoclonal Antibody Center, Washington State University) with its corresponding secondary antibody (Goat Anti-Mouse IgM, Human ads-PE/CY7, Southern Biotech, Birmingham, AL, USA), as detailed in Supplemental Table S1.In addition, each corresponding aliquot was labeled with an antibody pair to detect intracellular expression of total and phosphorylated (p) AKT, S6RP, and 4EBP1 (Cell Signaling Technology, Danvers, MA, USA) (Table S1).The antibodies used in the current experiment cross-react in bovine species, as documented by the manufacturer and previous studies (Mann et al., 2018;Sipka et al., 2020).Briefly, after thawing on ice, samples were first centrifuged for 10 min at 4°C, 500 × g, and the supernatant was discarded.Samples were then washed with 1 mL of 1 × PBS, separated into the 3 aliquots, and centrifuged for 10 min, at 4°C, 500 × g.The supernatant was discarded, leaving about 100 µL of 1 × PBS and cells.Then, samples were incubated for 1 h at room temperature, protected from light with the antibodies for the respective phosphorylated and total protein targets, as well as CD14 and CH138A.Samples were washed 2 times with 1 × PBS, as previously described, to remove unbound antibodies.The 100 µL of 1 × PBS and cells were incubated with PE/CY7, the secondary antibody for CH138A, for 15 min on ice.Samples were again washed 2 times, and cells were resuspended in 350 µL of 1 × PBS.Finally, cells were fixed with 150 µL of 4% paraformaldehyde (Cat.No. BP531-25, Fisher Scientific) and protected from light at 4°C until further analysis.Samples were measured in an Attune Acoustic Focusing Flow Cytometer (Thermo Fisher Scientific, Waltham, MA, USA) and analyzed using FlowJo V10 software (BD Biosciences, San Jose, CA, USA).Controls included unstained cells and cells with single stains for each antibody.Cell subpopulations were quantified as a percentage of total cells.

Liver biopsies, RNA extraction, and quantitative PCR analysis
Liver samples were taken via biopsy puncture from cows under local anesthesia (Dann et al., 2006) at −10 d and 1 d relative to subclinical mastitis challenge.The liver was frozen immediately in liquid nitrogen and stored at -80°C until further analysis.Total glutathione (GSH) and oxidized glutathione (GSSG) in liver tissue were measured with a commercial kit (Cat.No. NWH-GSH01; Northwest Life Science Specialties LLC, Vancouver, WA).Reduced GSH was calculated as reduced GSH = total GSH -GSSG.
Total RNA was extracted from liver tissue using Trizol reagent (Invitrogen, Cat.No. 15596018, USA), and RNeasy® Plus Mini Kit (Qiagen, Cat.No. 74134, Germany), following the manufacturer's instructions with some modifications.Briefly, 200mg of liver tissue immersed in 1 mL of Trizol was transferred to a 2 mL RNase-free O-ring tube, containing one 5 mm stainless steel bead (Qiagen, Cat.No. 69989, Germany) and homogenized in a Bead beater (Qiagen TissueLyser LT) for 1 min.After homogenization, the lysate was transferred to a 2 mL RNase/DNase-free microtube, and 200 µL of phenol: chloroform (Invitrogen, Cat.No. AM9730, USA) at 4°C was added to isolate the RNA from the organic phase.After centrifugation at 13,000 × g for 15 min at 4°C, the upper phase was transferred into a new 2 mL RNase-free microtube.RNA purification steps were performed using a fully automated QIAcube connect machine (QIAcube connect, Qiagen, Germany) according to the manufacturer's recommendation.The RNA quantity (899.41 ± 343.91 ng/μL; mean ± SD) and purity as 260/280 ratio (2.07 ± 0.03) were determined via NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies).The RNA quality and integrity were evaluated using the Qubit 4 fluorometer (Invitrogen, Cat.No. Q33238) Qubit RNA IQ assay kit protocol (Invitrogen, Cat.No. Q33222).The final RNA IQ number for all samples was 9 ± 1.1.
The complementary DNA (cDNA) synthesis was performed according to (Bionaz and Loor, 2007) with modifications.Each cDNA was synthesized by reversed transcription using 100 ng of RNA, 1 μL of Random The qPCR reaction was performed in a QuantStudio 6 Flex System Real-Time PCR (Applied Biosystems, USA) in MicroAmp ® Optical 384-well Reaction Plate (Applied Biosystems, Cat No. 4309849, USA) as described in (Bionaz and Loor, 2007).The genes glyceraldehyde 3-phosphate dehydrogenase (GAPDH), ubiquitously expressed transcript isoform 2 (UXT), and ribosomal protein 9 (RPS9) were used as internal control genes (ICG).Data were then normalized with the geometric mean of the 3 ICG.Details of selected genes, and designed primers evaluation are presented in the supplemental materials.Additional mechanistic information on target genes was obtained by calculating the percentage relative abundance of mRNA (Supplemental Table S4) (Bionaz and Loor, 2007).

Statistical analysis
The effects of Met on blood biomarkers were evaluated at 0, 6, 12, 24, 36, 48, and 72 h relative to the SMC.Data collected over time (i.e., performance data, blood biomarkers, PCR, blood flow cytometry) were evaluated by repeated measures using the MIXED procedure of SAS 9.4 (SAS Institute Cary NC, USA).The statistical model contained the effects of treatment (Trt), time (T; day or hours relative to SMC), and their interaction as fixed effects, while block and cow within treatment were considered as a random effect.Interactions with parity, baseline milk production, and DIM were tested and removed from the model when P > 0.20.Repeated measured data were modeled, selecting the variance-covariance structures with the least Bayesian information criterion value among compound symmetry, autoregressive 1 [AR(1)], or heterogeneous autoregressive 1 [ARH(1)].Single time point data were analyzed following the same model without the time statement.
Blood biomarker and flow cytometry data were logscale transformed if needed to comply with a normal distribution of residuals.Blood biomarker data from 0 to 72 h relative to SMC was unequally spaced; therefore, the spatial power covariance structure was used for this analysis.Statistical significance was declared at P ≤ 0.05 and tendencies at P ≤ 0.10.Observations were considered outliers when studentized residuals exceeded an absolute value of 4 and consequently excluded from the analysis.

Indicators of clinical disease
Results from health checks after the SMC are shown in Figure S3.No difference (P > 0.05) was observed between treatments.Clinical signs of mastitis, such as high temperatures, redness and hardening of the infected quarter, or abnormal milk, were not observed after the SMC.The lack of clinical signs confirmed the adequateness of the subclinical mastitis model used in this study.

Milk production and composition
The main effects and interactions for milk production and milk composition variables are presented in Table 2.There was a treatment × time (Trt × T) interaction for milk fat percentage (P = 0.01, 1E), milk fat yield (P = 0.01, Figure 1F), and a trend (P = 0.08) for milk protein percentage (Figure 1G) and milk ECM (P = 0.09, Figure 1C).The Trt × T interaction observed in milk fat percentage was mainly associated with a greater milk fat percentage in SM cows compared with CON at 24 h (P < 0.01) and 36 h (P = 0.05) relative to SMC, and a trend for the same effect at 48 h (P = 0.10) and 72 h (P = 0.07) relative to SMC.The latter was reflected in greater (P = 0.04) milk fat percentage in SMC compared with CON.The Trt × T in milk fat yield (kg/milking) was associated with greater milk fat yield in SM cows compared with CON at 24 h (P < 0.01) relative to SMC, and a trend for the same effect at 36 h (P = 0.09) and 72 h (P = 0.06) relative to SMC.This resulted in greater (P = 0.05) overall milk fat yield in SM cows compared with CON.The trend for the Trt × T observed in milk protein percentage was associated with greater (P ≤ 0.04) milk protein percentage in SM group compared with CON at 0, 60, and 72 h relative to SMC, and a trend for the same effect at 48 h (P = 0.09) relative to SMC.The latter was reflected in a trend for greater overall milk protein percentage in SM cows compared with CON.The trend for the Trt × T observed for ECM was associated with greater (P ≤ 0.03) ECM in SM cows compared with CON at 24, 36, and 72 h relative to SMC.This resulted in greater (P = 0.02) overall ECM in SM cows compared with CON.The SM cows had greater overall milk protein yield (P = 0.02) and total solids (P = 0.02) compared with CON.Similar to ECM, a trend for greater milk yield per day (P = 0.08;

Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS
Figure 1A) and per milking session (P = 0.07; Figure 1B) was observed in SM cows compared with CON.The SM cows produced ca.1.7 kg and 0.8 kg more milk per day and per milking session, respectively, than CON.The DMI, lactose percentage, MUN, and milk somatic cell linear score (MSCLS) were not affected by treatment effects (P > 0.10).

Blood biomarkers
Immunometabolism effects of SM were evaluated from 0 to 72 h relative to SMC and are present in Table 3.A Trt × T interaction was observed for Ca (P = 0.05, Figure S4).The Trt × T in Ca was mainly associated with a trend (P = 0.07) for lower Ca in SM cows compared with CON at 12 h relative to SMC.Lower ceruloplasmin (P < 0.01) and ROM (P = 0.01) were observed in SM cows compared with CON.Glucose, BHB, NEFA, D-lactic acid, L-lactic acid, albumin, haptoglobin, paraoxonase, globulin, FRAP, and MPO were not affected by Trt or Trt × T (P > 0.10).
Liver glutathione status.Greater overall total liver GSH (P = 0.05) was observed in SM cows compared with CON, while no Trt × T interaction was observed.In addition, no Trt × T or overall Trt effects were observed for GSSG and reduced GSH (P > 0.10).

Blood and milk leukocyte populations, L-selectin expression, and mTOR activity
Greater (P = 0.04) pAKT: AKT ratio was observed in isolated milk neutrophils in SM cows compared with CON at 24h relative to SMC (Table 4).Similarly, pS6RP: S6RP ratio was greater in neutrophils (P = 0.04) and monocytes (P = 0.01) in isolated milk cells from SM cows than CON.A trend (P = 0.08) for greater pAKT: AKT ratio was observed in milk leukocytes in SM cows compared with CON.In blood, a greater (P = 0.02) concentration of neutrophils was observed in SM cows compared with CON.Similar to blood neutrophils, the p4EBP1: 4EBP1 ratio in blood neutrophils was greater (P = 0.02) in SM cows compared with CON.Additionally, a trend (P = 0.07) for greater leukocytes L-selectin was observed in SM cows compared with CON.

Hepatic gene expression.
Met metabolism There was no Trt × T effect for any of the genes related to Met metabolism (Table 5).However, the overall expression of BHMT (P = 0.03) and PEMT (P = 0.02) were upregulated in SM cows compared with CON.Similarly, a trend for an mRNA upregulation (P = 0.09) of MAT1A was observed in SM cows compared with CON.No Trt effect (P > 0.10) was observed for the mRNA expression of SAHH, BHMT2, CBS, MTR, and CTH.
Glutathione Metabolism Overall, OPLAH was upregulated (P < 0.01) in SM cows compared with CON, while a trend (P = 0.08) for an upregulation GSS was observed in SM cows compared with CON.No Trt effect (P > 0.10) was observed for the expression of GPX1, GCLC, GGT, and GGCT.
Inflammatory response There was no Trt × T interaction for any of the genes related to inflammation.However, MYD88 and NRF2 were upregulated (P = 0.01) in SM cows compared with CON.In contrast to MYD88 and NRF2, HP was downregulated (P = 0.04) in SM cows compared with CON.The mRNA expression of TRAF6, NFKB1, NFKB1A, TNFA, TLR2, and CP were not affected by Trt (P ≥ 0.12).
Oxidative Stress There was no Trt × T interaction or overall Trt effect for any of the genes related to oxidative stress.

DISCUSSION
Mastitis has been identified as the most common and problematic disease in dairy cows.This disease negatively affects animal welfare and increases the use of antimicrobials, which are considered a cause of the onset and spread of antimicrobial resistance in the environment.The interest in preventing mastitis has been increasing in the last decades.Nutritional approaches are one of the most used preventive methods to enhance the animal's immune system and allow them to be more effective in combat infections (Hogan and Smith, 2012;da Costa et al., 2016;El-Sayed and Kamel, 2021).In our experiment, we used RPM in the form of SM while inducing subclinical mastitis to study the effects of RPM on lactation performance, inflammation, antioxidant activity, and pathways related to the immune system.

Milk production and composition
Increased milk yield has been commonly reported when supplementing RPM to peripartal (Sun et al., 2016;Zhou et al., 2016b) and mid-lactation (Wang et al., 2010;Junior et al., 2021) dairy cows.However, milk yield responses to RPM supplementation during a subclinical mastitis challenge have been seldom reported.In our study, we observed a greater ECM in SM cows compared with CON, which was rooted in both improvements in milk yield and components.While the trend for greater milk yield per milking session (Figure 1B) underscored transient improvements in milk yield, the daily milk yield (Figure 1A) data suggest a stronger positive effect of RPM supplementation on milk yield during a SMC.Adequate supply of AA to the mammary gland involves multiple factors, including the activity of several enzymes and transporters, dietary energy, and metabolizable protein (MP) (Bell et al., 2000;Cant et al., 2018;Huang et al., 2021).Enhancing the availability of essential AA (EAA), such as Met, through diet supplementation may, therefore, modify the EAA profile in MP, leading to improved absorption by the mammary gland.This adjustment could partly explain why SM cows tended to produce more milk than CON.Furthermore, Coleman et al. (2021) highlighted a link between the activation of the transsulfuration pathway via RPM supplementation and increased milk yield in transition dairy cows, associating this with enhanced liver function and reduced oxidative stress.Given the well-established transition cow model for inflammation and oxidative stress, it is plausible that these effects can be extrapolated to other adverse conditions in dairy cows, such as mastitis.Consequently, mitigating oxidative stress in the liver might account for the observed increase in milk production in SM cows during the SMC condition.This underscores the role of liver function in dairy cow resilience to disease.
A recent meta-analysis evaluating the effect of supplementing RPM to dairy cows on milk composition revealed that cows supplemented with RPM have greater milk fat percentage and milk protein percentage in comparison with control cows (Wei et al., 2022).Similarly, in previous studies, during stress conditions, such as the peripartal period and heat stress, RPM supplementation has increased or maintained milk fat percentage and milk fat yield (Osorio et al., 2013;Zhou et al., 2016b;Pate et al., 2020).The latter is in line with the present study, where cows supplemented with RPM had an increase in milk fat percentage and milk fat yield.It has been suggested that the effects of Met on milk fat may be related to the facilitated transport of lipids to the mammary gland via lipoprotein synthesis such as very-low density lipoproteins (VLDL) (Bauchart et al., 1998;Toledo et al., 2021).This could be partially ascribed to better liver function, maintaining adequate synthesis of apolipoproteins needed for VLDL assembly (Osorio et al., 2013;Sun et al., 2016).However, in vitro work has suggested that choline and not Met enhance VLDL export in neonatal calf hepatocytes (Chandler and White, 2017).The notable differences in milk fat, especially the increased levels in SM cows compared with CON at 24 and 36 h post-SMC (Figure 1E), may stem from the interaction between RPM supplementation effect and the cows' physiological responses to SMC conditions.However, it is important to note that our study's design limits our ability to separate the effects of RPM and SMC effects.
Greater milk protein percentage at 0 h relative to SMC confirms the positive effects of RPM supplementation commonly observed previously in healthy mid-lactation dairy cows (Junior et al., 2021;King et al., 2021).The overall greater milk protein yield in SM cows demonstrates a greater resilience in SM cows during a SMC.Similarly, others have observed a greater milk protein percentage and yield during stress conditions such as the peripartal period (Osorio et al., 2013;Zhou et al., 2016b) and heat stress (Pate et al., 2020).

Inflammation and oxidative stress
Ceruloplasmin is a multifaceted protein regulating metabolic balance of copper and iron, antioxidant activity, and is commonly characterized as an acute-phase protein (Liu et al., 2022).The latter has been well documented in transition dairy cows, where the blood concentration of this positive acute-phase protein increases during inflammatory conditions commonly observed soon after calving (Bertoni et al., 2008).Stress models have reported an increased blood ceruloplasmin concentration in early-lactation cows challenged with intramammary E. coli LPS (Minuti et al., 2015).A similar increase in ceruloplasmin (CP; gene symbol) mRNA expression has been observed in mammary gland tissues in cows naturally infected with coagulase-positive staphylococci (Zalewska et al., 2020).Even though it is considered that soluble ceruloplasmin is mainly synthesized in liver (Liu et al., 2022), the latter effects may suggests a localized effect of ceruloplasmin at the mammary gland level.The increased ceruloplasmin observed during mastitis conditions suggests a pivotal role for this acute-phase protein to bind and transport Cu and Fe, as well as antioxidant effects through oxidation of Cu and Fe (Liu et al., 2022).
In the current study, the lower ceruloplasmin in SM cows coupled with better lactation performance indicates a lesser need to mount a strong inflammatory response.This is corroborated by the high levels of plasma ceruloplasmin in CON cows, which were above a 2.7 µmol/L threshold indicated by Trevisi and Minuti (2018) for cows at risk of disease during the transition period.
Oxidative stress is the loss of reduction-oxidation (redox) homeostasis, leading to excess oxygen radicals (Holmstrom and Finkel, 2014).Oxidative stress is commonly regarded as an underlying factor for inflammation that can impair the immune system while triggering an amplification of pro-inflammatory pathways (Sordillo and Mavangira, 2014).Regarding mastitis, Laliotis et al. (2020) reported a correlation between oxidative stress biomarkers (e.g., ROS) at calving and clinical mastitis incidence in early lactation (75% of positive cows with <45 DIM).Others have reviewed this connection between redox imbalance, immune dysfunction, and mastitis (Khan et al., 2023).Therefore, it is conceivable that the high ceruloplasmin levels accompanied by the greater ROM in CON cows were conducive to detrimental effects on Glutathione serves as a potent reducing agent for the glutathione peroxidase defense mechanism against oxidative stress (Lu, 2013).Through the transsulfuration pathway, Met can supply the needed cysteine, which, along with glutamate and glycine, form the de novo GSH synthesis in liver (Lu, 2013).In transition dairy cows, previous studies have shown an increase in liver GSH when supplementing rumen-protected Met (Osorio et al., 2014b;Zhou et al., 2016a;Batistel et al., 2018).In the reduced form, GSH plays an important role in cellular metabolism and protects against free-radical-induced oxidant injury (Morand et al., 1997).Overall, we observed greater levels of GSH and reduced GSH in the liver of SM cows.Due to the link between mastitis and oxidative stress (Sordillo and Mavangira, 2014;Laliotis et al., 2020), the lower oxidative stress (low ROM) in SM cows could be partially explained by the greater GSH.

mTOR signaling and adhesion molecules
The PI3K/AKT/mTORC1 kinase pathway, also known as the mTOR pathway, stands out as the primary intracellular route for detecting nutrient balance.It has become recognized as a crucial mediator for incorporating metabolic signals into the activation of immune cells across different mammalian species (Weichhart et al., 2008).This was confirmed in the context of transition dairy cows, where Sipka et al. (2020) observed a role for mTOR signaling in LPS-activated bovine immune cells and a potential for a nutrient-enhanced response of targets of mTOR signaling.For instance, in PMN cells from transition cows activated with LPS, the p4EBP1: 4EBP1 ratio was enhanced when PMN cells were incubated with a nutrient mixture (i.e., AA, insulin, and glucose) in comparison with PMN incubated with PBS.Similarly, blood neutrophils from SM cows had a greater p4EBP1: 4EBP1 ratio than CON cows.Phosphorylated 4EBP1 is a result of an active mTORC1 signaling that occurs in response to abundant nutrients and energy.Although Met activation of mTORC1 activity is beyond the scope of this study, others have provided evidence that Met can maintain mTORC1 activity during oxidative stress conditions in mammary epithelial cells from sows (Zhong et al., 2021).The greater ratios of pAKT: AKT and pS6RP: S6RP in milk neutrophils suggest that Met may play a role in maintaining mTOR signaling even after neutrophils' transendothelial migration.However, this effect needs further confirmation.L-selectin is a cell adhesion molecule expressed on the surface of most leukocytes, such as lymphocytes, monocytes, and neutrophils.It plays a critical role in the immune system, particularly in mediating the migration of leukocytes to sites of inflammation or injury (Ivetic et al., 2019).Although it is conceivable that activation of mTOR via Met supplementation could lead to greater protein synthesis (e.g., translation) and, therefore, greater L-selectin synthesis, more causative connections between mTOR and L-selectin have been reported.For instance, Sinclair et al. (2008) observed that mRNA expression of L-selectin in T cells was regulated by the PI3K-mTOR axis via regulation of the transcription factor Kruppel-like factor 2 (KLF2).The stark differences in biological functions between T cells and neutrophils preclude a direct extrapolation of mTOR signaling influence in L-selectin expression in neutrophils in the current study.However, the authors speculate that a similar or alternative mTOR-related mechanism, as described by Sinclair et al. (2008) could have induced the greater Lselectin expression in blood neutrophils from SM cows compared with CON.

Transcriptional regulation in hepatic tissue
The mRNA expression of genes related to the Met cycle in hepatic tissue from transition cows has been previously evaluated (Osorio et al., 2014a;Zhou et al., 2017).The hepatic mRNA expression of MAT1A, PEMT, and SAHH was commonly upregulated when cows were supplemented with Met (Osorio et al., 2014a;Zhou et al., 2017).Consistent with those studies, we observed an upregulation of MAT1A and PEMT in SM cows when compared with CON (Table 5).This indicates that the Met cycle may rely on transcriptional upregulation of the 2 initial steps to respond to increased Met supply.In contrast to MAT1A and PEMT, BHMT was also downregulated in SM cows compared with CON, which deviates from a consistent lack of response in Zhou et al. (2017) and Osorio et al. (2014a).The latter effect suggests that during a SMC, any additional flux of homocysteine through the Met cycle is regenerated to Met via MTR rather than BHMT.In fact, Zhou et al. (2017) data indicate that MTR enzyme activity is more responsive to increased levels of Met than BHMT.
The last step in the synthesis of GSH is carried out by GSS, which binds glycine to the γ-glutamylcysteine complex performed by GCLC (Franklin et al., 2009).The 5-oxoprolinase is an intermediate enzyme in the glutathione metabolism that catalyzes the hydrolysis of 5-oxoproline to glutamate (Van der Werf et al., 1971).In the current study, we observed an upregulation of GSS and OPLAH in SM cows, which could partially explain the increased liver GSH concentration in SM cows.
Contrasting effects were observed on mRNA expression (Table 5) and blood protein concentration (Table 3) of ceruloplasmin and haptoglobin.While blood ceruloplasmin was lower in SM cows, CP mRNA expression was not different between SM cows and CON.In contrast, blood haptoglobin was not different between SM cows and CON, but HP mRNA expression was downregulated in SM cows.These inconsistencies might be a product of posttranslational modifications, rates of clearance and synthesis, and diurnal changes in mRNA abundance and blood concentration of these acute-phase proteins.However, the lower blood ceruloplasmin coupled with HP downregulation in SM cows suggests a lower local and systemic inflammatory condition in SM cows.

CONCLUSIONS
This study showed that supplementing rumen-protected methionine to early and mid-lactation dairy cows during a subclinical mastitis challenge maintains greater milk yield while improving milk protein and fat content.In terms of oxidative stress, SM cows had lower blood ROM concentration and greater liver glutathione.In addition, Met supplementation might attenuate inflammation and enhance blood and milk immune cells' protein synthesis through the mTOR pathway.However, further research is needed to understand these effects on immune cells, and gene expression of cytokines, transcription factors, and metabolic markers, among others.
Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS

Figure 1 .
Figure 1.Daily milk yield (A), milk yield per milking session (B), energy-corrected milk (C), milk somatic cell linear score (MSCLS) (D), milk fat percentage (E), milk fat yield (F), milk protein percentage (G), milk protein yield (H) from control cows (CON) and cows supplemented with rumen-protected methionine (SM) during a subclinical mastitis challenge.Values are means, with standard errors represented by vertical bars.
Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS

Table 2 .
Feed intake and lactation performance parameters in control cows (CON) and cows supplemented with rumen-protected methionine (SM) during a subclinical mastitis challenge Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS

Table 3 .
Paz et al.: SUPPLEMENTAL METHIONINE AND SUBCLINICAL MASTITIS Blood and liver biomarkers from 0 to 72 h relative to a subclinical mastitis challenge (SMC) for control cows (CON) and cows supplemented with rumen-protected methionine (SM)

Table 5 .
Hepatic gene expression of target genes related to methionine metabolism, glutathione metabolism, inflammatory response, and oxidative stress in control cows (CON) and cows supplemented with rumen-protected methionine (SM) at −10 and 1 d relative to a subclinical mastitis challenge