. Effect of close-up metabolizable protein supply on colostrum yield, composition, and immunoglobulin G concentration and associations with prepartum metabolic indicators of Holstein cows

The prepartum diet as well as individual metabolic status of the cow influences colostrum parameters. The objectives of this study were to 1) investigate the effect of increasing prepartum dietary MP supply on colostrum yield, composition, and immunoglobulin G (IgG) concentration, and 2) identify prepartum metabolic indicators associated with these outcomes. Multiparous Holstein cows (n = 96) were blocked by expected calving date and randomly assigned to 1 of 2 prepartum diets formulated to contain a control (CON; 85 g of MP/kg DM; 1,175 g of MP/d) or high (HI; 113 g of MP/kg DM; 1,603 g of MP/d) level of MP starting at 28 d before expected calving. Both prepartum diets were formulated to supply Met and Lys at an equal amount of 1.24 and 3.84 g/Mcal of metabolizable energy (ME), respectively. Metabolic indicators were determined in serum (albumin, glutamate dehydrogenase, cholesterol, aspartate transaminase, total protein, total bilirubin, and IgG) or plasma (Ca, glucose, fatty acids, BHB, and urea nitrogen) twice weekly in a subset of cows (n = 60). Colostrum was harvested at 3.6 ± 2.4 h from calving and yield as well as concentrations of IgG, fat, protein, and Ca were determined. Cows were retrospectively grouped based on the typical volume of colostrum needed for 2 colostrum meals (<6 or ≥ 6 kg), IgG concentration (<100 or ≥ 100 g/L), as well as the median concentrations of fat (<4.4 or ≥ 4.4%), protein (<16.5 or ≥ 16.5%), Ca (<0.21 or ≥ 0.21%), and total colostrum ME (<8.65 or ≥ 8.65 Mcal). Data were ana - lyzed using mixed effects ANOVA, with repeated measures where applicable. Feeding HI tended to increase colostrum yield in cows entering parity 2 (9.4 vs. 7.2 ± 0.9 kg), but treatment did not affect yield from cows entering parity ≥3 (5.1 vs. 6.4 ± 1.0 kg). Supply of MP did not affect concentrations of IgG, fat, protein, or Ca. Cows that produced ≥ 6 kg vs. those producing <6 kg of colostrum had lower plasma concentrations of glucose. Metabolic indicators were not associated with IgG group. Colostrum fat ≥4.4% was associated with cows having lower prepartum concentrations of glucose, total protein, albumin, and aspartate transaminase activity. Colostrum protein ≥ 16.5% was associated with lower circulating serum IgG and elevated cholesterol. Elevated glucose as well as lower cholesterol and BHB concentrations were associated with colostrum Ca ≥ 0.21%. Further, higher albumin and fatty acids as well as lower glucose concentrations were associated with a greater colostrum energy output. In conclusion, increasing prepartum MP supply tended to increase colostrum yield in cows entering parity 2, but did not affect the composition or IgG concentration. The observed associations between metabolic indicators and colostrum parameters suggest that slight adjustment in metabolism during late gestation might be necessary to support colostrogenesis, but the causality of these relationships should be considered.


INTRODUCTION
Colostrum has a rich nutritional profile and has been recognized for its immunological and bioactive components (Blum and Hammon, 2000, Mann et al., 2020, Chandler et al., 2023).Colostral components, such as fat, protein, and lactose, are nutritionally valuable for the newborn calf (Lopez and Heinrichs, 2022) and can be altered by the prepartum diet of the dam (Mann et al., 2016, Martinez et al., 2018).Several other factors including individual cow as well as dry period managerial factors have been identified as associated with colostrum production (reviewed by Westhoff et al., 2024a), emphasizing the complexity of colostrogenesis.
Prepartum dietary strategies can have a particular effect on colostrum yield and composition.Although research into nutritional interventions during the transition period has advanced our understanding of the effect of prepartum MP supply on milk production (Farahani et al., 2017, Cardoso et al., 2020, Zang et al., 2022), less is known about the effect of MP supply in the close-up period on colostrum yield, quality, and composition from dairy cattle.Increasing the availability of circulating AA could directly provide substrates for colostrogenesis or indirectly affect colostrum production by influencing development of mammary stromal and epithelial cells.Current research on this topic has yielded mixed results on the effect of MP on colostrum IgG concentration from multiparous cows (Akhtar et al., 2022, Van Hese et al., 2023).In addition, our understanding is hampered by the use of different model systems to estimate the prepartum MP supply, differences in handling the inclusion of firstlimiting AA, differences in base diets, or a general lack of MP estimation.
Colostrum synthesis aligns with a period of growing nutrient demands and reductions in DMI that alter the metabolic status during late gestation.Given these physiological adaptations, it is plausible that adjustments in maternal metabolism could influence substrates or signals that regulate colostrum synthesis.Prepartum metabolic indicators have been explored to understand the potential relationship of prepartum metabolism on colostrum outcomes.In Rossi et al. (2023), cows producing ≥6 kg of colostrum had elevated prepartum antioxidant potential, BHB concentrations, as well as a lower oxidant status index.Immler et al. (2021) associated elevated prepartum concentrations of calcium and glutamate dehydrogenase activity with lower colostrum Brix %.Moreover, colostrum yield has been associated with postpartum metabolic indicators, such as circulating fatty acid and BHB concentrations (Karl and Staufenbiel, 2016), as well herd-level hyperketonemia prevalence (Westhoff et al., 2023a).These studies collectively suggest a relationship between indicators of energy status and colostrum yield, but it remains unclear whether colostral components such as the concentration of fat, protein, and Ca are also associated with metabolic indicators.Further identification of metabolic indicators associated with colostrum parameters in different source populations are needed to strengthen the external validity of existing relationships as well as to understand substrates that could influence colostrogenesis.
As such, we investigated the effect of prepartum MP supply as well as associated indicators of prepartum metabolism with colostrum production from Holstein dairy cows.We hypothesized that prepartum MP supply during late gestation would affect colostrum parameters and that these parameters are associated with prepartum metabolic indicators.Our objectives were to 1) examine the effect of increasing prepartum MP supply on colostrum yield, composition, and IgG concentration and 2) associate prepartum metabolic indicators with the yield and composition of colostrum.

MATERIALS AND METHODS
All procedures were approved by the Cornell University Institutional Animal Care and Use Committee (protocol number 2019-0031).A detailed description of the study animals, diets, cow management, and milk production were described in our companion manuscript (Westhoff et al., 2024b).Briefly, multiparous Holstein cows (n = 96) were enrolled in a randomized block design between May and November 2021 at the Cornell University Ruminant Center.Cows were moved to individual tiestalls between 42 and 35 d before expected calving and fed a far-off diet.At 28 d before expected calving, cows were blocked into 24 groups by expected calving date and balanced for parity and previous lactation 305-d mature equivalent milk production.Animals were randomly assigned within block to ad libitum intake of a close-up TMR formulated to contain either a control (CON; 85 g of MP/kg of DM) or high (HI; 113 g of MP/kg of DM) level of MP.The concentration of MP was increased in the HI diet using heat-treated soybean meal (Amino Plus, Ag Processing Inc.) and a protein supplement (ProvAAl Lysine, Perdue Agribusiness).Rumen protected Met and Lys (Smartamine M, Adisseo Inc.; USA Lysine, Kemin Industries Inc.) were used to formulate both close-up diets to contain 1.24 and 3.84 g/Mcal of ME, respectively using the Cornell Net Carbohydrate and Protein System v. 6.5.5 (AMTS.Cattle.Professional v. 4.17.0.0;AMTS LLC; Van Amburgh et al., 2015).Forage and TMR sampling as well as analytical methods were described in detail in our companion manuscript (Westhoff et al., 2024b).In brief, composite samples of each forage ingredient as well as 4-wk composite samples of each TMR were submitted to a commercial laboratory (Cumberland Valley Analytical Services) for wet chemistry analysis and prediction of chemical composition by near-infrared reflectance spectroscopy and wet chemistry analysis of minerals, respectively.Results of the forage analysis were entered into AMTS to estimate the weekly MP and ME supply for each diet.

Animal Sampling and Analysis
At 0630 h and before morning feed delivery, blood samples were collected from the coccygeal vessels once before assignment to treatment and twice weekly from −28 d before expected calving until 3 DIM into plain collection tubes and those containing 158 USP units of Westhoff et al.: Effect of close-up metabolizable… sodium heparin (Becton Dickinson).Serum and heparin plasma were harvested by centrifugation at 2,300 x g at 4°C for 20 min, frozen, and stored at −80°C until analysis.On a random subset of 30 cows per treatment, prepartum serum concentrations of albumin, cholesterol, total protein, and total bilirubin as well as aspartate transferase (AST) and glutamate dehydrogenase (GLDH) activity were determined using commercial reagents (Catachem Inc.) on a biochemistry analyzer (CataChemWell-T; Catachem Inc.; Abuelo et al., 2020).Plasma total calcium was determined in samples collected −10, −6, and −3 d relative to calving at the Cornell University Animal Health and Diagnostic Center using an o-Cresolphthalein colorimetric assay.Serum IgG concentrations were determined for prepartum samples as well as at 3 DIM via radial immunodiffusion (RID; Triple J Farms) according to the manufacturer's instructions.Additionally, prepartum plasma concentrations of glucose, nonesterified fatty acids (NEFA), urea nitrogen (PUN), and BHB were determined on the same subset of cows in duplicate by enzymatic colorimetric analysis (PGO enzyme preparation, Millipore Sigma; HR Series NEFA-HR (2), Wako Life Sciences; urea nitrogen procedure number 640, Millipore Sigma; Chandler et al., 2022), respectively.Plasma concentrations of BHB were measured with samples warmed to 37°C using a Precision Xtra point-of-care device (Abbott) as previously described (Leal Yepes et al., 2018).The intra-and interassay CV were 2.9 and 7.0 percent for glucose, 4.4 and 9.4 percent for NEFA, and 5.1 and 5.9 percent for PUN, respectively.
Each cow was carefully monitored during parturition and the calf was separated from the cow immediately following calving by farm personnel.Colostrum was collected into a stainless-steel bucket in a double 12 parallel parlor (DeLaval) within 8 h of calving.Colostrum weight was recorded on a digital scale and a composite sample was used to determine Brix % on a digital Brix refractometer (Model PA201, Misco).A 50-mL composite colostrum sample was collected and subsequently aliquoted to either be frozen at −20°C for IgG, calcium, galactose, and lactose analysis or was diluted 2-fold with PBS, mixed with bronopol preservative, and stored at 4°C for analysis of fat and protein by Fourier Transform Infrared Spectroscopy (972.160;AOAC International, 2012) and SCC by optical fluorescence (method 972.160;AOAC International, 2012) at a commercial laboratory (Dairy One Cooperative Inc.).An additional aliquot of colostrum was submitted to the forage laboratory (Dairy One Cooperative Inc.) for calcium analysis (Sirois et al., 1994).Colostrum IgG concentration was determined on colostrum diluted 8-fold with sterile saline by RID (Triple J Farms) according to the manufacturer's instructions.A non-IgG protein fraction was calculated as the difference of protein and IgG concentration.Lactose and free galactose concentrations were determined on a random subset of 27 cows in each treatment group with a commercial assay (MAK017; Sigma-Aldrich) according to the manufacturer's instructions.Briefly, galactose was oxidized in whole colostrum diluted 640-fold with PBS to produce colorimetric product within the linear range of detection between 2 to 10 nmol.Inclusion of a sample blank as well as a sample with lactose hydrolyzed by lactase were included to determine free and total galactose, respectively.Lactose concentration was determined by subtracting free from total galactose.The intra-and interassay CV were 2.3 and 9.5%, respectively.Colostrum ME was determined via the NRC (2001) equation: where 0.97 and 0.96 were used to convert gross energy to digestible energy and digestible energy to ME, respectively.Energy concentration was multiplied by colostrum yield to determine the total colostrum energy output for each cow.

Analytical Approach
Sample size estimation was performed before the study as previously described (Westhoff et al., 2024b).Variables were created to dichotomize colostrum yield and Brix % as < 6 and ≥ 6 kg as the volume of colostrum needed for 2 colostrum meals (Westhoff et al., 2023b) and < 22.0 and ≥ 22.0% as the cutpoint associated with highquality colostrum, respectively.Because all samples had an IgG concentration ≥ 50 g/L, IgG concentration was categorized as < 100 and ≥ 100 g/L (Mann et al., 2016).Chi-squared tests were used to investigate differences in calf sex as well as categorized colostrum yield, IgG, and Brix % using PROC FREQ (SAS v. 9.4; SAS Institute Inc.).Treatment differences in the outcome variables of calf birth weight, dry period length, colostrum yield as well as colostrum IgG, ME, fat, protein, calcium, galactose, lactose, total solids, and linear score were determined using mixed effects ANOVA (PROC MIXED; SAS v. 9.4; SAS Institute Inc.).Effects of treatment, parity (entering 2 vs. ≥ 3), and the interaction of treatment and parity were included as fixed effects and enrollment block was included as a random effect.Model assumptions of normality and homoscedasticity of the residuals were assessed visually.
To determine the association between metabolic indicators and the yield as well as IgG, fat, protein, calcium, and total ME content of colostrum, cows were retrospectively categorized based on their colostrum composition.Colostrum yield and IgG concentration were categorized as < 6 and ≥ 6 kg and < 100 and ≥ 100 g/L as described above.Colostrum fat, protein, calcium, and total ME were dichotomized using the median.Repeated measures ANOVA with the fixed effects of treatment, group, time, and parity (entering 2 vs. ≥ 3) were performed in PROC MIXED with an antedependence covariance structure.Enrollment block was included as a random effect and time was included as a repeated effect.Model assumptions of normality and homoscedasticity of the residuals were assessed visually.Glutamate dehydrogenase, AST, and NEFA were transformed using the natural logarithm to meet the model assumptions.Differences in serum and colostrum IgG concentration as well as yield by parity were investigated on a subset of cows (n = 60) using mixed effects ANOVA with repeated measures when applicable with the effects of parity (2 vs. ≥ 3), treatment, and time.The relationship between serum IgG concentration and the concentration and yield of IgG in colostrum was assessed with Pearson correlation.Data are reported as LSM and SEM or back-transformed LSM and 95% confidence intervals.Significance was declared when P ≤ 0.05 and a tendency was declared when P ≤ 0.10.

RESULTS
Colostrum samples from 3 cows were not collected (HI: 2; CON: 1); therefore, samples from 93 cows were included in the analysis of colostrum yield and composition.Cows were dry for an average ± SD (range) of 60.4 ± 4.5 (50 to 74) d and on the close-up treatment diets for 27.4 ± 4.4 (17 to 39) d, and there were no differences by treatment (P > 0.24).Prepartum DMI as a percent of BW did not differ for HI and CON (1.80 vs. 1.77 ± 0.03% BW; P = 0.29), respectively.Calf birth weight (female: 42.3 vs. 42.7 ± 0.8 kg; male: 44.6 vs. 46.1 ± 0.8 kg; P ≥ 0.23) and sex (HI: female: 18, male: 28, twins: 0; CON: female: 20, male: 26, twins: 1; P = 0.56) also did not differ for HI and CON, respectively.Colostrum was harvested on average 3.6 ± 2.4 h from calving and did not differ by treatment (P = 0.26).A detailed description of diet ingredients and formulation was previously described (Westhoff et al., 2024b).Table 1 shows the analyzed composition of prepartum diets.Predicted MP supply was 1,603 g/d for HI and 1,175 g/d for CON.

Associations between metabolic indicators and colostrum production
The median (range) total bilirubin was 0.05 (0.00 to 0.44) mg/dL and total bilirubin was not detected in 251 of 503 samples (49.9%); therefore, total bilirubin was excluded from statistical analysis.The association between metabolic indicators, colostrum yield, and IgG groups are shown in Tables 3 and 4, respectively.Cows that produced ≥ 6 kg of colostrum had lower plasma concentrations of glucose (65.5 vs. 67.9± 0.9 mg/dL; P < 0.01) compared with cows that produced < 6 kg, respectively.Albumin, GLDH activity, cholesterol, AST activity, total protein, IgG, Ca, NEFA, BHB, and PUN did not differ by colostrum yield group (P ≥ 0.15).Metabolic indicators were not associated with IgG group (P ≥ 0.14).
Mean (SD) serum IgG during the close-up period was 30.1 (11.0) g/L.The mean (quartile 1, quartile 3) intracow coefficient of variation for serum IgG concentration during the close-up period was 28.8 (21.6, 37.6) %.The inter-cow coefficient of variation for the mean serum IgG concentration during the close-up period was 22.9%.The concentration of IgG in serum was greater in parity ≥ 3 (32.8± 1.0 g/L) compared with parity 2 (28.4 ± 0.9 g/L; P < 0.01; Figure 1).Colostrum IgG concentration was higher in parity ≥ 3 (144.4± 7.5 g/L) compared with parity 2 (98.6 ± 6.9 g/L; P < 0.01) but the yield of IgG in colostrum (720 vs. 658 ± 67.9 g/L; P = 0.45) and colostrum yield (5.8 vs. 7.4 ± 0.7 kg; P = 0.12) did not differ for parity ≥ 3 and 2, respectively.Beyond a moderate correlation (r = 0.41; P = 0.03; Figure 2) between colostrum IgG concentration in parity ≥ 3 cows and the change in serum IgG concentration from −32 d relative to calving to the minimum concentration, the mean, slope, and change in serum IgG concentrations were not associated with colostrum IgG concentration or yield (P ≥ 0.09).

Effect of 2 MP levels on colostrum yield and composition
The first objective of this study was to determine the effect of increasing prepartum MP supply on colostrum yield and composition.The proportion of cows in the current study producing enough colostrum to feed a calf at least 2 colostrum meals (≥6 kg, 49.5%) was higher than the 34.7 and 40.0%previously reported (Rossi et al., 2023, Westhoff et al., 2023b).Our study found a tendency that prepartum MP supply increased colostrum yield from cows entering parity 2, but no effect was identified for higher parities.Colostrum yield from multiparous Holstein cows did not differ when increasing prepartum MP supply, estimated using NRC (2001), from 744 to 976 and 849 to 1,200 to 1,387 g of estimated MP/d (Farahani et al., 2017, Farahani et al., 2019).Furthermore, feeding 2 levels of MP prepartum (65 vs. 90 g/kg DM) at 2 levels of DMI to achieve NRC (2001) estimated MP supply of 593, 823, 902, and 1,258 g/d did not affect colostrum yield (Akhtar et al., 2022).Although colostrum yield was not affected by total MP supply in the aforementioned studies, the interaction between MP supply and parity was not tested by the respective authors.It is plausible that our finding of higher colostrum yield in cows entering parity 2 was directly supported by AA as substrates for colostrum synthesis or that increased circulating AA indirectly affected colostrum production by providing glucose and energy precursors for the liver and mammary gland or by supporting stromal and epithelial development.We currently lack understanding of an optimum dietary metabolizable AA supply for prepartum cows and whether increasing circulating AA during late gestation could alter metabolic pathways or tissue development that might affect the process of colostrogeneis.Because of these knowledge gaps and the high individual variability in colostrum yield (Borchardt et al., 2022, Westhoff et al., 2023b) replication of these data, including data from nulliparous heifers, are needed to confirm this finding.
In a study by Van Hese et al. (2023), a treatment x parity interaction was observed such that cows entering parity 2 fed an elevated MP supply, estimated with the DVE-system in the Dutch energy and protein system (Tamminga et al., 1994), during the far-off (1,203 vs. 846 g of estimated MP/d) and close-up (1,631 vs. 1,258 g of estimated MP/d) periods produced colostrum with greater IgG concentration (61.3 ± 2.3 vs. 55.2 ± 2.8 g/L).In contrast to the aforementioned study, we did not observe a treatment x parity interaction for the concentration of IgG in colostrum.The small absolute difference of 6 g/L in colostrum IgG concentration observed in the treatment x parity interaction in the study by Van Hese et al. ( 2023) is likely of low biological significance.Although we hypothesized that altering MP supply could affect IgG or component concentrations by directly affecting the transfer capacity or synthesis of colostrum constituents, by altering colostrum yield, or a combination, the current study did not support an effect of MP supply on colostrum composition.In agreement with our findings,  increasing the crude protein concentration (Santos et al., 2001, Toghyani and Moharrery, 2015, Westhoff et al., 2023a), MP concentration (Akhtar et al., 2022), or inclusion of rumen protected Lys (RPL; 0.54% RPL of dietary DMI; Fehlberg et al., 2020)  been shown to alter lipogenic gene networks, miRNA, and fatty acid synthase expression involved in the regulation of fat synthesis in bovine mammary epithelial cells (Dudek andSemenkovich, 1995, Li et al., 2016).In a study by Chandler et al. (2022), intravenous infusion of a mixture of AA to early lactation cows increased yields of de novo and preformed fatty acids in milk.Although adipose tissue mobilization likely contributed to elevated yields of preformed fatty acids in the aforementioned study, the authors discussed a contribution of AA to the concurrently observed increase in de novo synthesis of FA in AA infused cows (Chandler et al., 2022).Because

Associations between metabolic indicators and colostrum production
Our second objective was to identify prepartum metabolic indictors associated with the yield and composition of colostrum.In contrast to the current study which did not identify an association between colostrum parameters and GLDH activity or serum Ca, elevated prepartum Ca concentrations and GLDH activity were associated with a lower colostrum Brix % (Immler et al., 2021).Moreover, cows producing colostrum with an IgG concentration ≥ 50 g/L were associated with elevated prepartum concentrations of glucose and albumin compared with those producing < 50 g of IgG/L (Rossi et al., 2023), a relationship that was not observed in the current work.
Interestingly, we observed remarkable within and between cow variability for serum concentrations of IgG during late gestation, but beyond a moderate correlation between colostrum IgG concentration and the change in serum IgG from −32 d to the minimum concentration for the individual cow in parity ≥ 3, we did not find meaningful associations between the change or concentration of IgG in circulation and the yield or concentration of IgG in colostrum.Colostral IgG are thought to be transported from maternal circulation into the mammary gland by a specific Fc receptor on the mammary gland (Fc receptor of the neonate; FcRn) that is expressed in acinar epithelial cells of the mammary gland during late gestation (Mayer et al., 2005).The lack of a strong relationship between circulating and colostral IgG in our study invites a closer look at the role of the FcRn.First, the FcRn appears to have a role in recycling high affinity molecules, such as IgG2, back to circulation (Cianga et al., 1999) and likely contributes to IgG half-life and the greater concentra-tion of IgG1 compared with IgG2 in bovine colostrum although they are in equal concentration in circulation (Hurley and Theil, 2011).Using transgenic mice for the overexpression of bovine FcRn in the mammary gland, Lu et al. (2007) observed increased concentrations of IgG in serum and milk compared with control mice.In addition, when the transgenic mice were injected with an equal amount of bovine IgG1 and IgG2, the IgG1 serum to milk ratio did not differ between transgenic and control mice but the ratio of IgG2 in serum to milk was greater in transgenic mice (Lu et al., 2007) providing evidence that the expression of FcRn might affect IgG recycling of high-affinity proteins such as IgG2 back into circulation while selectively extracting IgG1.We did not characterize the proportion of the different IgG subtypes in circulation and in colostrum in this study.Second, recent data from other species challenges the presumed sole origin of colostral IgG from circulation by FcRN mediated transfer.Using FcRn knockout pigs, Ke et al. (2021) revealed that despite lower serum and colostrum IgG concentrations in knockout compared with wild type pigs, the ratio of serum IgG to colostrum IgG concentration did not differ between groups.This suggests that FcRn was not solely responsible for IgG transport into the mammary gland.Future work is needed to understand the timing and factors that affect the expression of FcRn in late gestation dairy cattle and to what extent the expression of FcRn influences IgG accumulation in the mammary gland and recycling.
Our results show that production of colostrum with total ME ≥ 8.65 Mcal was associated with elevated prepartum NEFA concentrations, whereas higher colostrum yield, fat, and total ME were associated with lower concentrations of glucose during the prepartum period.We have previously associated NEFA concentrations ≥ 290 µEq/L determined between 3 to 14 d before expected calving with an elevated colostrum yield (Westhoff et al., 2023a).Further, Rossi et al. (2023) observed elevated BHB and lower cholesterol in cows producing ≥ 6 kg of colostrum.Because colostrogenesis is a nutrient intensive process (Soufleri et al., 2021), the lower glucose as well as elevated fatty acid or ketone concentrations during late gestation suggest a greater energy demand to support colostrum synthesis.However, future investigations are needed to determine the causality of these relationships as it is plausible that elevated circulating fatty acids could supply the mammary gland with necessary substrates to support colostrum synthesis.
The activity of GLDH and AST were determined in the current study as possible indicators of hepatocellular integrity.In an observational study by Immler et al. (2021), GLDH activity, determined on average 8.2 d before calving, had a negative correlation with colostrum Brix %, but no association was observed between AST and Brix %.Here, neither AST nor GLDH activity were associated with IgG group, but an association was observed with colostrum fat.Aspartate aminotransferase and GLDH activity are greater during early lactation compared with the dry period (Seifi et al., 2007, Stoldt et al., 2015, Andjelić et al., 2022), but have not been associated with concentrations of milk fat during lactation (Mordak et al., 2020, Andjelić et al., 2022).While results from our study suggest an association between energy metabolites and colostrum production, colostrum synthesis appeared to have minimal associations with prepartum indicators of liver health.

CONCLUSIONS
Increasing the MP supply during the close-up period tended to increase the yield of colostrum from cows entering parity 2 but not from parity ≥ 3 cows, and Westhoff et al.: Effect of close-up metabolizable… Westhoff et al.: Effect of close-up metabolizable… Westhoff et al.: Effect of close-up metabolizable… Figure 1.Serum IgG concentration by parity group (2: n = 32, ≥ 3: n = 28).

Table 1 .
Westhoff et al.: Effect of close-up metabolizable… Analyzed composition of far-off and close-up diets

Table 2 .
Colostrum yield, composition, and IgG concentration from cows fed diets differing in MP concentration from −28 d before expected calving until calving

Table 3 .
Metabolic indicators from prepartum Holstein cows associated with colostrum yield group or rumen protected Met (RPM) and RPL (0.09% RPM and 0.15% RML of dietary DMI; O'Meara et al., 2023) in prepartum diets were not associated with colostral composition or IgG concentrations.While the concentration of colostrum fat was not affected by MP supply in the present study, small differences in colostrum yield and fat concentration resulted in a difference in fat yield.Essential AA have Westhoff et al.: Effect of close-up metabolizable…

Table 4 .
Metabolic indicators from prepartum Holstein cows associated with colostrum IgG group

Table 5 .
Metabolic indicators from prepartum Holstein cows associated with colostrum fat group

Table 6 .
Metabolic indicators from prepartum Holstein cows associated with colostrum protein group Westhoff et al.: Effect of close-up metabolizable…

Table 7 .
Metabolic indicators from prepartum Holstein cows associated with colostrum calcium group

Table 8 .
Metabolic indicators from prepartum Holstein cows associated with colostrum total metabolizable energy (ME) group