Plant polyphenol extract supplementation affects performance, welfare and the Nrf2-oxidative stress response in adipose of heat stressed dairy cows

We examined the effects of a supplement of plant polyphenols extracts of green tea, capsicum and fenugreek, and electrolytes [(Na + , K+), AXT ; Axion ThermoPlus, CCPA, France] during summer heat load on production, welfare, and oxidative stress proteins in adipose tissue ( AT ) of dairy cows. Forty-two multiparous mid-lactation cows were divided into 3 groups during summer, and were fed for 2 wks either a standard milking cows' diet ( CTL , n = 14), or supplemented with 100 g/d of AXT ( 100AXT , n = 14), or 150 g/d of AXT ( 150AXT , n = 14), while being cooled 5 times a day; then, half of the cows from each dietary treatment were cooled ( CL ) or not-cooled ( NCL ) for 2 wks, after which the CL/NCL were switched for additional 2 wks. Cows were milked 3 times a day and milk composition was analyzed at the end of each 2 wk period. Vaginal temperature ( VT ) was measured for 3 consecutive days in each period. Biopsies of subcutaneous AT were taken from 10 NCL cows (5 CTL and 5 150AXT) at the end of the period, and examined by LC-MS/MS proteomics analysis. Data were analyzed with PROC MIXED of SAS; the model included the effects of dietary treat-ment, cooling regimen, period, and their interactions.


INTRODUCTION
Environmental heat load due to elevated ambient temperature and humidity has adverse effects on the physiology of livestock, eliciting a reduction in feed intake and production, and increased inflammation and oxidative stress (Filipe et al., 2020).The increased oxidative stress can activate antioxidant pathways to counteract the negative effects of heat load due to excessive production of reactive oxygen species (ROS).In dairy cows, high levels of catalase (CAT), superoxide dismutase (SOD), glutathione reductase and malondialdehyde (MDA) have been observed in summer months (Yatoo et al., 2014;Zachut et al., 2020).Heat load also promotes gut leakiness, and it was proposed that the combination of an altered rumen environment and a "leakier" gut under heat stress might contribute to lipopolysaccharide (LPS)-induced inflammation (Pearce et al., 2013;Bradford et al., 2015).During chronic heat load in summer, cows in late pregnancy were shown to have increased systemic levels of oxidative stress and inflammation compared with cows calving in winter, and their subcutaneous adipose tissue (AT) proteome Plant polyphenol extract supplementation affects performance, welfare and the Nrf2-oxidative stress response in adipose of heat stressed dairy cows was enriched with proteins related to the nuclear factor erythroid 2-related factor 2 (Nrf2)-mediated oxidative stress response (Zachut et al., 2017).Thus, nutritional supplementation that can alleviate oxidative stress may have desirable effects on heat-stressed dairy cows.
Many plant derived polyphenols are known for their anti-oxidative properties, and several studies have examined the effects of supplementation of plant polyphenols to ruminants.Plant polyphenols can serve as antioxidants by reducing ROS (Amorati et al., 2013), and they have been reported to increase endogenous antioxidants, including SOD, CAT, and glutathione (GSH; Masella et al., 2005).As reviewed by Oh et al. (2017a), some studies have reported that plant polyphones had positive effects on productivity, and in non-ruminants, it was demonstrated that plant polyphenols such as eugenol, cinnamaldehyde, and capsaicin, activate mammalian transient receptor potential (TRP) channels that are expressed on neurons, the intestines, the pancreas, immune cells, and other tissues (Vennekens et al., 2008;Holzer, 2011).Their binding to TRP channels activates signaling pathways within the cells and may regulate pro-or anti-inflammatory responses and oxidative stress in non-ruminants (Zeng et al., 2015).
Green tea (Camellia sinensis L.) is a source of polyphenols, mainly catechins (Crespy and Williamson, 2004), but also contains saponins, caffeine, and l-theanine.Green tea extracts possess antioxidant, antimicrobial, anticoccidial, and antiprotozoal activities (Heleno et al., 2015;Kolling et al., 2018).Recent studies examined the effects of supplementation of green tea extracts in dairy cows.Gessner et al. (2020) supplemented green tea extracts (10 g/d) to transition cows, and found that it reduced the expression of unfolded protein response genes in liver, suggesting a reduction in ER stress; however, other inflammatory indices were not altered by treatment.Others reported that green tea extract supplementation (5 g/d) to transition cows reduced ROS levels in erythrocytes (Vizzotto et al., 2021).Ma et al. (2021) supplemented transition cows with 0.2 g/ kg DM of green tea polyphenols, and reported that hyperketonemic cows supplemented with green tea had lower plasma concentrations of ROS, MDA, and hydrogen peroxide, and greater activities of SOD, glutathione peroxidase, and total antioxidant capacity.Moreover, in vitro studies have demonstrated that green tea polyphenols had beneficial effects on redox balance in bovine mammary epithelial cells (Ma et al., 2018), by reducing cellular oxidative stress via the inhibition of ROS accumulation.It was proposed that this effect can be partly explained by the activation of the Nrf2/heme oxygenase-1 (HMOX1) pathway (Ma et al., 2019).Oh et al. (2017b) supplemented dairy cows with ru-men protected Capsicum oleoresin and challenged them with LPS; they observed that cows supplemented with capsicum tended to have lower cortisol levels 2 h post LPS, and had lower levels of haptoglobin and thiobarturic acid reactive substances in plasma 24 h post challenge compared with controls.This demonstrated that capsicum supplementation can modulate acute phase responses induced by bacterial infection in dairy cows.Hence, plant derived polyphenols could potentially have beneficial effects on oxidative stress, however there is scarce information on the effects of plant polyphenols during heat load.Therefore, the objective of the present study was to examine the effects of a plant polyphenol extract supplement containing green tea, capsicum and fenugreek and electrolytes (Na + , K+) on performance, welfare indices and oxidative stress proteins in the AT of mid-lactation dairy cows during heat load.As we hypothesized that the supplementation will improve the response of cows to heat stress, we included in our experimental model a period of increased heat load (no cooling sessions), to examine the effects when cows are under intense heat stress conditions.

Animals and experimental procedures
The experimental protocol for the study was approved by the Volcani Center Animal Care Committee (approval number 903/21 IL), and the experiment was performed in accordance with ARRIVE guidelines and regulations.The study was conducted at the Volcani Institute experimental farm in Rishon Lezion, Israel.Forty-two multiparous (parity 2-5) mid-lactation (180 ± 61 DIM) cows were divided according to milk yield, parity, BW and days in lactation into 3 groups during peak summer in Israel (August-September), and were fed for 6 wks either a standard milking cows' diet (CTL, n = 14), or supplemented with 100 g/d of AXT (100AXT, n = 14), or 150 g/d of AXT (150AXT, n = 14), while being cooled 5 times a day.After 2 weeks of adjustment to dietary supplementation (period 0), half of the cows from each dietary treatment group were cooled (CL; 5 cooling session a day) and half were not cooled (NCL) for 2 wks (period 1), after which the CL/ NCL were switched within in each dietary treatment for additional 2 wks (period 2).A diagram describing the experimental model is presented in Figure 1.Supplements were top dressed individually and mixed with the upper third of the TMR daily at 11:00.The AXT supplement contained green tea, capsicum and fenugreek, as well as electrolytes (19.2% Na + , 3% K+, 9.6% Ca, 0.1% Mg and 0.2% Cl) and bicarbonate as a buffer.All cows were group-housed in a covered pen with an adjacent outside yard, and fed ad libitum once a day with a common Israeli diet that contained 16.5% crude protein and 1.78 Mcal net energy for lactation (NE L ) per kg dry matter.Individual feed intake was monitored by an automated system.The ingredients and chemical composition of the experimental diet is presented in Table 1.Cooled cows were exposed to 5 cooling sessions a day in the holding area of the milking parlor; the cooling sessions began at 04:10, 10:15, 12:10, 17:00, and 19:40 h.Each cooling session lasted 45 min, comprising repeated cycles of 30 s of showering and 4.5 min of forced ventilation without showering, as described in Honig et al. (2012).All cows were milked 3 times daily at 05:00, 13:00, and 20:00 h, and milk yields were recorded electronically at each milking, and also weighed automatically after each milking with a walk-in electronic scale (S.A.E.Afikim, Kibbutz Afikim, Israel).Milk production and BW were recorded electronically 3 times daily in the milking parlor (SAE, Kibbutz Afikim, Israel).In addition, milk samples were collected every 2 weeks, at the end of each session, from 3 consecutive milking, and analyzed for milk fat, protein, lactose, and urea by infrared analysis (standard IDF 141C:2000) at the laboratories of the Israeli Cattle Breeders' Association (Caesarea, Israel).Ambient temperature and relative humidity for thermal humidity index (THI) calculations (Dikmen et al., 2008) were recorded within the experimental barn by a system that included 2 sensors installed at both sides of the barn (Smaxtec Base station and climate sensors, Smart Farm Data, Hamilton, NZ).The average THI in the barn during the first period of CL/NCL was 80.1, and the average THI during the second period of CL/NCL (switch) was 78.4 (SEM = 0.75; P = 0.10).The cows were equipped with collar-mounted tags (HR-Tags; SCR Engineers Ltd., Hadarim, Netanya, Israel) that monitored and transmitted rumination time.Rumination data were recorded by a special microphone that detected chewing actions by analyzing vocal signals.Data were stored in 2-h blocks and uploaded 3 times daily at the milking parlor.The cows also were equipped with another sensor (Pedometer Plus; S.A.E.Afikim) that monitored lying time (min).The tag was fitted to the rear leg of each cow and the data were accumulated and transmitted to management software (AfiFarm; S.A.E.Afikim).Respiratory rate (RR) and rectal temperature (RT) were recorded twice weekly immediately after the noon milking at 14:00 h.The RT was recorded by a clinical thermometer inserted into the rectum for 1 min.The RR was measured by counting flank movements during 1 min.Vaginal tem- During the first 2 wk, cows were supplemented while being cooled 5 times a day (period 0); then, half of each dietary group were cooled (CL) and half were not cooled (NCL) for 2 weeks (period 1), after which the CL/NCL were switched for 2 weeks (period 2).Dietary supplementation groups were constant during the experiment.Created with Biorender.com.perature (VT) was measured in 8 randomly selected cows from each dietary treatment (4 CL and 4 NCL in each treatment) by temperature loggers every 10 min for 3 consecutive days in 24 cows in each period.Blood samples were collected weekly during the study from the coccygeal vein into vacuum tubes containing lithium heparin (Becton Dickinson Systems, Cowley, England).The blood samples were collected after the noon milking at 14:00 h, and plasma was separated after centrifugation at 4000g for 15 min and then stored at −80°C pending analyses.

Analysis of NEFA, oxidative stress and inflammatory markers in blood plasma
Plasma samples were analyzed for concentrations of NEFA, MDA, cortisol, haptoglobin (HP) and LPSbinding protein (LBP).Plasma NEFA concentration was determined using the NEFA C Test Kit (Wako Chemicals GmbH, Neuss, Germany).Plasma MDA concentration was measured by the fluorometric thiobarbituric acid reactive substances assay (Feldman, 2004).Cortisol concentrations were determined by ELISA (EIA1887, DRG International, Inc., Springfield, NJ, USA).Plasma HP level was determined by bovine ELISA kit (E-10HPT, ICL, Portland, OR).Concentrations of the acute-phase protein LBP were analyzed by a bovine LBP ELISA kit (OKCA02041, Aviva System Biology, San Diego, CA).

AT biopsies, mRNA abundance and immunoblot analyses
Subcutaneous AT biopsies were obtained from 10 randomly selected non-cooled cows at last day of the first 2-week period of NCL.Due to the invasiveness of the procedure, we were limited by the ethics committee to perform biopsies on 10 cows; as we estimated that the higher dose of AXT will be more likely to affect the AT, we conducted the biopsies on 5 CTL and 5 150AXT NCL cows.The AT samples were taken from the subcutaneous fat pad around the pin bones as previously described (Zachut et al., 2013), immediately frozen in liquid nitrogen, and stored at −80°C.
The abundance of mRNA of genes related to oxidative stress were measured in AT.The RNeasy lipid tissue micro kit was used to homogenize using metal bead to 40 mg of AT samples in 1 mL of lysis solution for RNA extraction (Qiagen, Hilden, Germany).The RNA purity was assessed using a Nanodrop, and the 260/280 ratio of the RNA quality was found to be greater than 1.85.A cDNA reverse transcription kit was used to make first-strand cDNA (Applied Biosystems, Foster City, CA).Real-time PCR was used to detect specific mRNA transcripts quantitatively using a StepOnePlus equipment (Applied Biosystems) and the SYBR green PCR mix (Invitrogen, Carlsbad, CA).We investigated at the expression of oxidative stress-related genes: heat shock protein family A1 70 KDa (HSPA1A), ubiquitinconjugating enzyme E2 K (UBE2K), glutathione Stransferase Mu 1 (GSTM1), LBP, the plasma form of glutathione peroxidase (GPx), stress induced phosphoprotein-1 (STIP1), transient receptor potential vanilloid 1 (TRPV1), hemopexin (HPX), and glutathione S-transferase M3 (GSTM3).The primers were listed in Supplementary Table 1.Data were standardized for the quantity of the reference gene GAPDH mRNA in AT samples, and primers were validated before use.After investigating several candidate genes in AT (BRPS2, GAPDH, and actin) using Normfinder software GAP-DH was chosen as a reference gene.The delta-delta CT (relative quantity, RQ) of each gene was employed for statistical analysis.

Sample preparation for proteomic analysis, liquid chromatography mass spectrometry
The protein extracts of the AT samples (n = 5 from CTL and 150AXT) were lysed and digested with trypsin using the S-trap method.The resulting peptides were analyzed using nanoflow liquid chromatography (nanoAcquity) coupled to high resolution, high mass accuracy mass spectrometry (Q HF).Each sample was analyzed on the instrument separately in a random order in discovery mode.

Data processing and bioinformatics
Raw data was processed with MaxQuant v1.6.6.0.The data was searched with the Andromeda search engine against the bovine proteome database appended with common lab protein contaminants.Quantification was based on the LFQ method, based on unique/all peptides.
Bioinformatics was conducted with Ingenuity Pathway Analysis (IPA) software (Qiagen), to examine the locations, function, top canonical pathways and networks according to the differential proteome in AT of AXT150 vs. controls.

Statistical analysis
Continuous variables (milk, milk solids, DMI, efficiency variables, plasma indices, rectal and vaginal temperatures, and respiratory rate) were analyzed as repeated measurements with the MIXED Procedure of SAS (version 9.2, SAS Institute., Cary, NC, 2002), according to the following model: where μ = the overall mean, T i = the fixed effect of treatment ( i = 1 to 3), R j = the fixed effect of parity (j = 2 or > 2), C(T) ik = the random effect of cow k nested within treatment j , D l = the fixed effect of DIM, R m = the fixed effect of period (1 or 2), G n = the fixed effect of cooling (cooled or non-cooled), R m × G n = the fixed effect of the interaction between cooling and period, and E ijklmn = the residual error.When relevant, variables were analyzed with the specific data of the pretreatment period as covariates.
The interactions dietary treatment × period, and dietary treatment × cooling regimen, were included in the model, but were found not significant (P > 0.20), and therefore excluded from the model.The autoregressive order 1 (AR 1) was used as a covariance structure in the model because it resulted in the lowest Bayesian information criterion (BIC) for most of the variables that were tested.
Proteomic data was analyzed by unpaired 2-tailed Student's t-test after log 2 transformation.Protein and gene abundances in AT were analyzed by unpaired 2-tailed Student's t-test.Significance was set at P ≤ 0.05 and tendencies were set at P ≤ 0.10.

Cow performance and circulating levels of metabolic and inflammatory markers
Data was analyzed for the effect of nutritional supplementation, cooling regimen, period, and the interactions between them.As shown in Table 2, the average milk production was higher in 100AXT than in controls (P = 0.04) and the effect of cooling on increasing milk yield tended to be significant (P = 0.09).Milk components were not different among groups, but the effect of period was significant for fat percentage (P < 0.0001) and lactose percentage (P = 0.02; Table 2).The fat percentage (3.96 vs. 3.60%, SEM = 0.09, P < 0.0001) and lactose percentage (4.96 vs. 4.86%, SEM = 0.03, P = 0.02) were higher in period 1 than in period 2. The 4% FCM and ECM yields were higher in 100AXT than Daddam et al.: PLANT POLYPHENOLS AFFECT HEAT STRESSED COWS in controls (P = 0.003), and the effects of cooling regimen (P = 0.02), period (P = 0.04) and the interaction cooling × period (P = 0.02) were significant for both 4% FCM and ECM (Table 2).The 4% FCM and ECM were lower in NCL vs. CL cows (P < 0.0001), and lower in period 1 than in period 2 (P = 0.01); in NCL cows, the 4% FCM and ECM were lower in period 1 than in period 2 (P = 0.002).
The DMI was higher in 100AXT compared with controls (P = 0.01), while 150AXT had intermediate values but was not different from controls.The effect of cooling regimen was significant for DMI (P = 0.001), as well as the interaction cooling × period (P = 0.01); the effect of period tended to be significant (P = 0.10; Table 2).The DMI of CL was higher than NCL cows (30.9 vs. 27.0kg/d, SEM = 0.41, P < 0.0001), and in NCL cows the DMI was lower in period 1 than in period 2 (P < 0.0001).The calculated EB and efficiency measures were not different among dietary groups, but the effect of cooling (P = 0.04; Table 2) was significant, in which the calculated EB of CL was higher than NCL cows (13.5 vs. 10.0Mcal/d, SEM = 0.45, P < 0.0001).
Only for cortisol, the effect of period was significant; the average concentrations of cortisol were lower in the second compared with the first period (17.2 vs. 28.5 ng/ml in second and first periods, respectively, SEM = 1.97,P < 0.0001).

Indices of heat stress and welfare in cows supplemented with plant extracts in summer
As shown in Table 3, the average rectal and vaginal temperatures were not different among dietary groups.The effect of cooling regimen was significant for RT (P = 0.03) and VT (P < 0.0001), the effect of period was significant for VT (P < 0.0001), and the interaction period × cooling was significant for VT (P = 0.0002; Table 3).The VT was higher in NCL vs. CL cows (39.8 vs. 38.9°C,SEM = 0.06, P < 0.0001), and higher in period 1 than in period 2 (39.6 vs. 39.2°C,SEM = 0.06, P = 0.0005), as the VT of NCL cows in period 1 was higher than the VT of NCL cows in period 2 (P < 0.0001).Interestingly, the proportion of time during the day that cows were with VT >39°C was lowest in 100AXT compared with controls (P < 0.0001), and also lower in 150AXT than in controls (P < 0.0001; Table 3).The effects of cooling (P < 0.0001) and of period (P = 0.02) on time with VT > 39°C were significant as well (Table 3).
No difference was observed in average respiration rates among groups (Table 3), while rumination time was higher in 150AXT than in control (P = 0.05; Table 3).Lying time was higher both in 100AXT (P = 0.05) and in 150AXT (P = 0.03) compared with the control, the effect of cooling regimen was significant (P = 0.02), as well as the effect of period (P < 0.0001) and the interaction period × cooling (P = 0.006; Table 3).The lying time was higher in CL than in NCL cows (588.2Values different at P ≤ 0.05; 1 100AXT = cows supplemented with 100 g/d of supplement containing plant extract and electrolytes; 2 150AXT = cows supplemented with 150 g/d of supplement containing plant extract and electrolytes; 3 FCM = fat corrected milk; 4 ECM = energy corrected milk; 5 Energy balance calculated according to NRC (2001) equations.vs. 564.2min/d, SEM = 6.4,P = 0.009), and was lower in period 1 than in period 2 (P < 0.0001).The NCL cows in period 1 tended to have lower lying time than in period 2 (P = 0.06), and the CL cows had lower lying time in period 1 than in period 2 (P < 0.0001).

mRNA and protein abundances in the AT
We have analyzed the mRNA and protein abundances related to oxidative stress, as well as LBP and TRPV1 in AT (Table 4).The expression of LBP tended to be higher in 150AXT than in control AT (P = 0.08), with no differences in the expression of other genes between groups (HSPA1A, UBE2K, GSTM1, GPx, STIP1, TRPV1, HPX, GSTM3; Table 4).The average protein abundances of TRPV1 (P = 0.04) and of LBP (P = 0.04) were higher in AT of 150AXT than in control (Table 5 and Figure 2), while other proteins were not different between groups (GSTM1, STIP1, MEK1, PON1; Table 5).

Proteomics of AT
In total, we identified and quantified 2,395 proteins in AT, from which 16 were differential between 150AXT and control AT (P ≤ 0.05 and fold change ± 1.5; Table 6).As shown in Figure 3, in AT of 150AXT we found increased abundances of peroxidasin [fold change (FC) = 1.6, P = 0.05], microsomal glutathione S-transferase 2 (FC = 2.5, P = 0.05) and heme oxygenase 1 (FC = 3.6, P = 0.03) compared with controls.From the quantified proteins, IPA software (Qiagen) identified 2036 proteins, from which 1231 were defined as located in the cytoplasm (62%), 192 in the extracellular space (10%), 330 in the nucleus (16%) and 247 proteins were located in the plasma membrane (12%, Figure 4A).The proteins were defined mostly to enzymes followed by transporter and transcription regulator molecular functions as demonstrated in Figure 4B.
cantly changed peptides including ZNFAND2B, VPS16, PXDN, HIKESHI, GRB2, HMOX1, MGST2, CD34 and SLITRK5 were represented in the molecular network related to Nrf2 mediated oxidative stress response in 150AXT vs Control AT (Figure 6).The network associated with the differential proteome in 150AXT AT showed HMOX1 inhibiting STAT1 and growth factor receptor-bound protein 2 (GRB2) inhibiting SPRR1B, SPRR2F regulated protein network related to Nrf2 mediated oxidative stress mechanism.

DISCUSSION
In the present study, supplementation of AXT at a rate of 100 g/d to cows during summer heat load positively affected production performance (milk, FCM 4% and ECM yields), and increased feed intake and lying time compared with controls.Moreover, the proportion of time that the cows were with vaginal temperature > 39°C during the day was lowest in 100AXT.Supplementing cows with AXT at 150 g/d increased rumination time and lying time, and these cows also had less time with vaginal temperature > 39°C compared  with controls.In the AT of NCL cows, 150AXT tended to increase mRNA abundance of LBP, increased the protein abundance of LBP and TRPV1, and the differential AT proteome was related to several canonical pathways such as acute phase response signaling and Nrf2-mediated oxidative stress response.Together, this work demonstrates that dietary supplementation containing plant polyphenols from green tea, capsicum and fenugreen extracts as well as electrolytes affects the performance and welfare of heat stressed mid-lactation dairy cows.In addition, this work highlights the nutriproteomic effects of this supplementation on subcutaneous AT.Supplementation of AXT at 100 g/d (100AXT) increased milk production by 8.3% and feed intake by 8.0% compared with the control cows.The negative effects of heat load on feed intake were proposed to be mediated by body temperature, and it was suggested that minimizing the increase in body temperature could improve nutrient intake (West, 1999;Honig et al., 2012); thus, the lower proportion of time with VT > 39°C in 100AXT cows might be related to the higher DMI in these cows.Decreased feed intake caused by heat load is one of the main causes of decreased milk yield (Collier et al., 1982;West, 2003), therefore, the increased feed intake is probably the reason for the increased milk production in the 100AXT group.In Kolling et al. (2018), cows were supplemented with green tea at 0.028% of DM, and had an increase in the digestible fraction of the ingested DM compared with controls; however, since we did not examine the digestibility of the diet in the present study, we cannot at-tribute the increased intake to changes in digestibility.In contrast to our findings, studies in which cows were supplemented with anti-oxidants during heat stress (Abeyta et al., 2023), with green tea extract during the transition period (Gessner et al., 2020;Vizzotto et al., 2021), or with capsicum in mid-lactation (Oh et al., 2017b) demonstrated no effect on intake or milk yield.On the other hand, supplementing cows with grape seed and grape marc meal extract rich in polyphenols was shown to increase milk yield (Gessner et al., 2015;Olagaray et al., 2019), and Ma et al. (2021) reported that green tea polyphenol supplementation increased milk yield in hyperketonemic postpartum cows.Based on several studies in ruminants, it was concluded by Oh et al. (2017a) that phytonutrients may have a positive effect on productivity, although rumen fermentation is not affected; they proposed that this effect could be attributed to energy partitioning for milk production through their effect on insulin secretion and sensitivity.In the present study, we did not examine plasma insulin concentrations, therefore we cannot validate this premise.Taken together, most anti-oxidant supplementations to dairy cows do not affect feed intake and production, but possibly the presence of specific plant polyphenols in the AXT supplement and the interaction between them encouraged feed intake; however, this issue requires further investigation.The increase in feed intake and milk production in the 100AXT group suggests that this is the optimal dose of this supplement to achieve improved production during heat load (compared with 150AXT).In the present study, the interaction period × cooling was significant for DMI, ECM and 4% FCM.We assume that the main reason for this effect was the tendency for lower THI in period 2 compared with period 1, which could have affected the response to cooling.
The AXT supplement also contained electrolytes, that could have potentially affected the cows' performance.According to the actual feed intake in the study, we calculated that the basal diet provided 250 g/d of Ca, 200 g/d of K, and 49 g/d of Na.The AXT supple- ment added 10-15 g/d of Ca (in 100AXT and 150AXT, respectively), 3-4.5 g/d of K (in 100AXT and 150AXT, respectively), and 19.2-28.8g/d Na (in 100AXT and 150AXT, respectively) to the diet.Thus, the relative supplementation of Ca and K by AXT was not pronounced, but the addition of Na by AXT (39-59% addition of Na relative to the controls in 100AXT and 150AXT, respectively) may have had an effect relative to the control diet.It was previously demonstrated in heat-stressed cows, that milk yield and DMI increased when Na increased from basal (0.18% Na, dry basis) to 0.55% dietary Na (Schneider et al., 1986); thus addition of Na by AXT could be related to the increased intake.
In the present study, the average concentrations of blood indices of stress, inflammation and oxidative stress were not different among dietary treatments.Plasma cortisol concentrations were lower in the second vs. first period regardless of dietary treatment.As the THI tended to be lower in the second vs. first period, we attribute the effects of period in this study mainly to the lower THI levels; thus, the reduced heat load could be associated with the lower levels of cortisol in those cows.Indeed, Zachut et al. (2017) showed higher plasma cortisol in cows that were under summer heat load compared with cows in the winter season.Recently, Abeyta et al. (2023) reported that supplementation of anti-oxidants to heat-stressed cows, as described above, decreased plasma insulin concentrations, and increased NEFA compared with controls.In addition, plasma concentrations of LBP were increased by heat stress, mainly due to a 64% increase in its concentration in heat-stressed cows that were supplemented with the antioxidant compared with thermoneutral controls (Abeyta et al., 2023).Others demonstrated that supplementing green tea extracts (10 g/d) to transition cows did not affect inflammatory indices (Gessner et al., 2020); however green tea extract supplementation (5 g/d) to transition cows reduced ROS levels in erythrocytes (Vizzotto et al., 2021).In hyperketonemtic transition cows supplemented with 0.2 g/kg DM of green tea polyphenols, lower plasma concentrations of ROS, MDA, and hydrogen peroxide, and greater activities of SOD, glutathione peroxidase, and total antioxidant capacity were reported (Ma et al., 2021).As mentioned above, in LPS-challenged cows, supplementation of capsicum lowered the levels of HP, suggesting an effect on acute phase proteins (Oh et al., 2017b).Together, it seems that dietary anti-oxidant supplementation does not affect blood indices of stress or inflammation in mid-lactation cows during heat load, but may have a stronger effect in transition cows that are more sensitive to stress and inflammation, or in cows facing an immune challenge (such as bacterial infection).
We found that welfare indicators were improved in cows supplemented with AXT during heat load; 100AXT increased lying time, and 150AXT increased rumination time and lying time compared with controls.The elevated lying time could be related to the lower VT threshold that we observed, in which both 100AXT and 150AXT had less proportion of time dur- ing the day with VT > 39°C, since cows with lower VT are more comfortable and may spend more time lying.The increased rumination time that we observed in 150AXT and numerically higher in 100AXT might be related to the increased lying time in the AXTsupplemented cows.The effect of the supplementation on these welfare parameters is unique, as most studies do not report of such effects when providing cows with anti-oxidants during heat load; for example, Abeyta et al. (2023) found no effect of anti-oxidant supplementation on rectal, vaginal and skin temperature, nor on respiration rate in HS or thermoneutral cows.This is the first study, to the best of our knowledge, to describe nutri-proteomic changes in subcutaneous AT of cows supplemented with plant polyphenols and electrolytes during heat load.Due to limitation in the number of cows that were approved to be biopsied, we were only able to examine AT from control vs. 150AXT non-cooled cows.We found several changes in protein and mRNA abundances in the AT, which could imply a local effect of the AXT supplementation within this tissue.The mRNA and protein abundance of LBP, an acute phase protein, was higher in AT of 150AXT than in controls.LBP binds to LPS of gram-negative bacteria and interacts with CD14, and this complex is capable of stimulating macrophages through toll-like receptor (TLR) 4 interaction, activating a pro-inflammatory response (Ceciliani et al., 2012).It was proposed that low abundance of AT LBP is related to an increase in inflammatory tone in bovine AT (Zachut and Contreras, 2022), thus the increase in LBP in 150AXT AT could indicate a lower inflammatory state.Furthermore, 2 of the top enriched canonical pathways in AT according to the differential proteome in 150AXT were acute phase signaling and LPS/IL-1 mediated inhibition of RXR function, which could support the effect of AXT supplementation on inflammatory processes in AT.However, Gessner et al. (2020) supplemented transition cows with green tea extract (20 g/d) and did not find differences in gene expression of pro-inflammatory, acute phase reaction or antioxidant genes in the liver; nonetheless, they observed an increase in expression of UPR gene in liver of cows supplemented with green tea, which could suggest an effect on ER stress.When green tea extract was supplemented to transition Jersey cows, a reduction in ROS in plasma and in erythrocytes and an increase in concentration of reduced gluthathione was observed (Vizzotto et al., 2021).Oh et al. (2017) suggested that supplementing phytonutrients reduces oxidative stress by decreasing lipid peroxidation and increasing endogenous antioxidants in ruminants.However, we did not find differences in oxidative stress indices in plasma between dietary groups, which may imply of a local effect within the AT rather than a systemic effect of this dietary supplementation.Possibly, the lower oxidative stress in AT of 150AXT is related to the lower inflammatory tone in their AT, as oxidative stress fuels inflammation (Zachut and Contreras, 2022).
Our proteomic analysis revealed increased abundance of peroxidasin (PXDN), microsomal glutathione S-transferase 2 (MGST2) and heme oxygenase 1 (HMOX1) in AT of 150AXT vs. controls.Moreover, the differential proteome in AT was enriched with pathways related to anti-oxidative pathways such as the Nrf2-oxidative stress response and glutathione redox reactions.The AT activates antioxidant defenses, such as the GSH, SOD, and the Nrf2 systems, and mitochondrial uncoupling, to neutralize free radicals that may be increased due to HS.It was demonstrated in bovine adipocytes that AMPK activation increases the antioxidant response by enhancing Nrf2 response through its key component Nrf2 and its downstream targets heme oxygenase 1 (HMOX1), SOD1, CAT, and glutathione-S-transferase (GST; Xu et al., 2021).The role of Nrf2 and oxidative stress in AT development and function is complex and depends on many factors, such as the various sources of oxidative stress, the temporal aspects of Nrf2 expression in AT differentiation, and the age and genetic background of the animal (Schnider and Chan, 2013;Zachut and Contreras 2022).As mentioned, proteomic analysis of AT from late pregnant cows during summer heat load highlighted Nrf2-mediated oxidative stress response as one of the most enriched pathways in AT compared with those calving in winter season (Zachut et al., 2017).Intriguingly, nutritional supplementation can alter Nrf2 activity; as methionine supplementa-tion reduced abundance of the Nrf2 protein in AT of postpartum cows compared with controls (Liang et al., 2019).Together, our data supports the premise that supplementation of specific plant derived anti-oxidants affects oxidative stress response elements in the AT, possibly as part of a defense mechanism during summer heat stress.
The abundance of TRPV1, a receptor sensitive to capsicum and to heat, was increased in AT of 150AXT compared with controls.The heat-sensitive TRPV1 is a ligand-gated ion channel that plays a key role in modulation of the sensation of pain and thermal hyperalgesia (Gavva et al., 2005).TRPV1 has been shown to be expressed in the AT (Christie et al., 2018), and both thermal heat (Christie et al., 2018) and capsicum (Caterina et al., 1997), among other activators, activate the TRPV1 receptor.Recently, we demonstrated that the mRNA abundance of TRPV1 was lower in AT of postpartum cows calving in summer heat load compared with winter, and the protein abundance of TRPV1 tended to be lower in AT of mid-lactation cows that were not cooled vs. cooled cows in summer (Kra et al., 2022).We propose that possibly the activation of the TRPV1 receptor both by capsicum from the feed supplement, and perhaps by the environmental heat load, may be part of the etiology explaining the effects of this supplementation on heat stressed cows.

CONCLUSIONS
Supplementation of AXT, which contains plant polyphenol extracts from green tea, capsicum and fenugreek as well as electrolytes, to heat stressed dairy cows increased DMI, milk and 4% FCM production, lowered the proportion of time with VT > 39°C and improved welfare indices.Most of the effects were evident at the dose of 100 g/d.In NCL cows, the AXT supplementation enriched the AT proteome with Nrf2-oxidaitve stress response and acute phase response proteins.Together, these findings suggest that plant polyphenols could exert positive effects on heat stressed cows, possibly via improving the oxidative stress response in tissues, as demonstrated by the nutri-proteomic effects in AT.Further large studies should be conducted to validate the effects of this supplementation on heat stressed dairy cows.

Figure 1 .
Figure1.Experimental model.Mid-lactation cows during summer were divided into 3 groups and fed for 6 weeks either a standard diet (control), or supplemented with 100 g/d of plant extract and electrolytes (100AXT), or 150 g/d of AXT (150AXT).During the first 2 wk, cows were supplemented while being cooled 5 times a day (period 0); then, half of each dietary group were cooled (CL) and half were not cooled (NCL) for 2 weeks (period 1), after which the CL/NCL were switched for 2 weeks (period 2).Dietary supplementation groups were constant during the experiment.Created with Biorender.com.
Figure 3. Protein abundances in AT according to proteomic data.Adipose tissue biopsies were collected from 5 control and 5 150AXT (150 g/d AXT; plant extract and electrolytes) non-cooled cows and analyzed by LC-MS/MS.
Figure 4. Location (A) and molecular function (B) of peptides identified by proteomic analysis in adipose tissues of 150AXT (150 g/d AXT; plant extract and electrolytes) vs. control according to Ingenuity Pathway Analysis (Qiagen).
Figure 5. Top canonical pathways in adipose tissues of 150AXT (150 g/d AXT; plant extract and electrolytes) vs. control, according to proteomic analysis and Ingenuity Pathway Analysis (Qiagen).
Figure 6.Molecular Network of Nrf2 mediated oxidative stress mechanism regulating molecules in 150AXT (150 g/d AXT; plant extract and electrolytes) vs. control adipose tissues according to proteomic analysis and Ingenuity Pathway Analysis (Qiagen).

Table 2 .
Milk yield, components, feed intake, energy balance and efficiency calculations in cows supplemented with plant extract and electrolytes during summer

Table 3 .
Daddam et al.:PLANT POLYPHENOLS AFFECT HEAT STRESSED COWS Indices of heat stress and welfare in cows supplemented with plant extracts and electrolytes in summer

Table 4 .
mRNA abundance in adipose tissue of cows supplemented with plant extract and electrolytes in summer

Table 5 .
Protein abundance in adipose tissue of cows supplemented with plant extract and electrolytes in summer

Table 6 .
Differential protein abundances according to proteomics of adipose tissue in cows supplemented with plant extract and electrolytes in summer