Lactational performance effects of supplemental histidine in dairy cows: A meta-analysis

The objective of this meta-analysis was to examine the effects of supplemental His on lactational performance, plasma His concentration and efficiency of utilization of digestible His (Eff His ) in dairy cows. The meta-analysis was performed on data from 17 studies published in peer-reviewed journals between 1999 and 2022. Five publications reported data from 2 separate experiments, which were included in the analyses as separate studies, therefore resulting in a total of 22 studies. In 10 studies, His was supplemented as rumen-protected (RP) His; in 1 study, 2 basal diets with different dHis levels were fed; and in the remaining experiments, free His was infused into the abomasum (4 studies), the jugular vein (3 studies) or deleted from a mixture of postruminally infused AA (4 studies). The main forages in the diets were corn silage in 14 and grass silage in 8 studies. If not reported in the publications, the supplies of dietary CP, metabolizable protein (MP), net energy of lactation, and digestible His (dHis) were estimated using NRC (2001). An initial meta-analysis was performed to test the standard mean difference (SMD; raw mean difference of treatment and control means divided by the pooled standard deviation of the means), that is, effect size, and the corresponding 95% confidence interval (CI) in production parameters between His-supplemented groups versus control. Further, regression analyses were also conducted to examine and compare the relationships between several response variables and dHis supply. Across studies, His supplementation increased plasma His concentration (SMD = 1.39; 95% CI: 1.17–1.61), as well as DMI (SMD = 0.240; 95% CI: 0.051–0.429) and milk yield (MY; SMD = 0.667; 95% CI: 0.468–0.866), respectively. Further, milk true protein


INTRODUCTION
Formulating diets to include less dietary CP and instead focusing on supplying adequate amounts of digestible EAA (dEAA) has become an increasingly important part of dairy cow nutrition.This practice is essential both in terms of decreasing N emissions to the environment and improving production and feed resource-use efficiency.The importance of increasing N efficiency of dairy cows by fully utilizing their ability to rely on microbial protein as the major source of dEAA as well as complementing the microbial dEAA with AA from the RUP fraction has been long acknowledged (Virtanen, 1966;Schwab and Broderick, 2017).
Histidine is a critical EAA in low CP dairy rations (Lee et al., 2012), where the proportion of microbial CP (MCP) contribution to dietary MP increases, acknowledging the low concentration of His in MCP (NRC, 2001).However, the establishment of His requirements has been challenging.Histidine is a unique AA in that it has labile pools in the body, including muscle dipetides (carnosine and anserine) and hemoglobin, that can be mobilized under His shortages and mask the true level of His deficiency (Giallongo et al., 2017;Lapierre et al., 2021).For example, it has been estimated that hemoglobin can supply 0.4 to 0.7 g/d of His over a 6 to 8-wk period in continuous experiments with basal diets supplying around 57 and 49 g/d dHis, respectively (Giallongo et al., 2016;Giallongo et al., 2017), or up to 2.3 g/d in a short-term crossover experiment with a basal diet supplying 31 to 33 g/d dHis (Lapierre et al., 2021).
Histidine as a limiting AA was first demonstrated in grass silage-based, low CP diets with postruminal infusion of His (Kim et al., 1999;Vanhatalo et al., 1999).More recent studies have further shown the importance of His in MP-deficient (MPD; Lee et al., 2012;Räisänen et al., 2021a), and MP-adequate (MPA), according to NRC (2001), corn silage-based diets (Giallongo et al., 2017;Räisänen et al., 2021bRäisänen et al., , 2022) ) with rumenprotected His (RPHis).Infusion of His into the abomasum of lactating dairy cows fed grass silage-based diet increased milk yield (MY) or milk true protein yield (MTPY; e.g., Vanhatalo et al., 1999;Korhonen et al., 2000;Huhtanen et al., 2002).Similarly, deletion of His from a mixture of infused AA decreased MY and MTPY (Kim et al., 1999;Hadrová et al., 2012;Lapierre et al., 2021).In the deletion studies, incremental His doses resulted in an even greater response in MTPY compared with other forms of His supplementation due to the greater His deficiency achieved in these experiments (Lapierre et al., 2021).
Overall, current data indicates that His may be limiting AA for milk and milk protein secretion in lactating dairy cows, but the extent of production responses to His supplementation may depend on various factors including type of basal diet and method of His supplementation, MP supply, as well as the supply and efficiency of utilization of other AA (Lapierre et al., 2021;Räisänen et al., 2021a,b).
Therefore, the objective of the current study was to conduct a meta-analysis to investigate the lactational performance of dairy cows receiving supplemental His and potential interactions with dietary factors and nutrient supply.The hypothesis was that His supplementation enhances DMI, MY, ECM yield, and MTPY of lactating dairy cows, and that the production responses to His supplementation are more pronounced with MPD diets.

Data Set
The data set for this meta-analysis was compiled using data obtained by searching the Scopus database using search terms "Histidine" and "Dairy cow".The search was limited by document type "Article" and resulted in 140 document results, wherefrom the data set was compiled.The data set included published peerreviewed journal articles in English that investigated the effect of His supplementation on milk production and composition variables and plasma His concentration in lactating dairy cows.Seventeen peer-reviewed publications with a total of 22 studies (including 5 publications reporting 2 substudies) were selected for this meta-analysis and are detailed in Table 1.In all experiments, feed was offered ad libitum, except in Lapierre et al. (2021), where cows were offered 97% of their previous ad libitum intake.The control treatments for the current analysis were selected so that they only differed from the respective His treatments by lack of His supplementation and were not confounded by other factors such as other EAA, MP or energy supply.

Calculations
Milk protein concentration was reported as CP in 6 of the publications.For these studies, milk true protein concentration was calculated as 0.95 × milk CP % (DePeters and Cant, 1992).Unless included in the publication, milk component yields were calculated by multiplying MY with component percentage, and ECM yield was calculated based on the reported MY and reported or calculated milk component composition according to Sjaunja et al. (1990).
Supplies of MP, dEAA and NE L were estimated by entering dietary and cow information (breed, BW, DIM, MY and milk composition) in NRC ( 2001) model for studies where these were not reported (7 publications).For the supply of dAA from infusion or rumen-protected AA (RPAA), the values given in each publication for ruminal outflow or amount of dAA supply from supplemental AA were used.For all studies, dEAA flow of NRC (2001) was corrected using correction factors from Lapierre et al. (2019) to account for incomplete recovery of AA with 24-h hydrolysis of proteins.Further, dEAA flow was corrected to remove the contribution from the duodenal endogenous flow, based on NASEM (2021).This adjusted net supply of dHis (named below as adjusted: adHis) was thereafter used in the analysis for the whole data set.The efficiency of utilization of His (Eff His ) was calculated as: Eff His = sum of His in true exported proteins (g) ÷ [adHis flow (g) -endogenous urinary loss (g)], where the sum of His in true exported proteins included MTPY and estimated scurf and metabolic fecal protein calculated as detailed in Lapierre et al. (2021).Lastly, marginal recovery of His in milk protein was calculated as (His in milk protein, g/d ÷ additional His supply, g/d) × 100.Concentration of His in MTPY was assumed to be 2.92% (Lapierre et al., 2020) and additional His supply (g/d) was the difference in adHis supply between the His-supplemented treatment and the control.

Statistical Analyses
Meta-Analysis.To examine the overall differences between His-supplemented and control groups standardized mean differences (SMD; SMD = raw mean difference of treatment and control means divided by the pooled standard deviation of the means; i.e., effect size, and the corresponding 95% confidence interval (CI) for DMI, MY, ECM, milk components, and plasma His) were computed using Comprehensive Meta-Analysis software (v.3; Biostat USA).A sensitivity test was conducted on the compiled data set to assess potential bias by using the "One study removed" function in Comprehensive Meta-Analysis.The data were assessed for the overall effect of His supplementation, and the studies were then grouped by MP-level into MPA (n = 13) or MPD (n = 9) studies, where MP-deficiency was defined as MP supply at or below 95% of NRC (2001) requirements, and the effect of His supplementation was investigated for each group separately.The effect size for DMI, MY, ECM, milk components and plasma His was calculated.Heterogeneity among studies was measured using Chi-squared (Q) test.Further, forest plots, assuming a random effects model, displaying the SMD at 95% CI were constructed to visually assess the effect of His supplementation on DMI, MY, and MTPY.The effect of His supplementation was deemed significant at P ≤ 0.05 and a trend at 0.05 < P ≤ 0.10 and SMD values of <0.2, 0.2 to 0.7, and >0.7, were considered small, moderate or large, respectively (Cohen 1988).
Regression Analysis.The relationship of the effect of adHis supply with the responses of DMI, MY, ECM, and MTPY were assessed assuming a nonlinear log logistic model form: , where y i and x i are the response and adHis supply values from ith study, respectively, and the output parameters represent the maximum response (M), the lower asymptote when adHis supply is low (D), rate of curve increase (K) and level of His at 50% of maximum response (L).This model describes an S-shaped curve which begins at a lower response level D, then increases to an inflection point at L, where the curve then slows and approaches the upper response level M. The log logistic model for these responses was selected based on the Aikake's comparison criteria (AICc; smaller is better) when compared the model fit of an increasing plateaued exponential model.The model comparisons for logistic model fit with AICc are given in Supplemental Table S1 (https: / / scholarsphere .psu.edu/resources/ c2bfa303 -70bd -4297 -8fb3 -e7399271e938; Räisänen et al., 2023).
The relationship of the responses of adHis supply, additional His supply or Eff His on Eff His , and marginal recovery, respectively, were assumed to follow a simple exponential model of the form: , where y i and x i are the response and adHis levels from the ith study, respectively, and the parameters represent an intercept, a, and a slope, b, (i.e., rate of change).This model starts at an initial level of the response at adHis = 0, a, and then gradually descends toward zero at high levels of adHis.All nonlinear models were weighted by the respective sample size of each study.Lastly, the relationships of adHis supply between plasma His concentration, or between adHis supply, the ratio of dHis to NE L , and Eff Met , Eff Lys , or Eff Leu and Eff His were tested with a simple linear models fit.
In each model, the random effect of study was included to account for study-to study variation (St-Pierre, 2001).Further, a random effect of treatment within study was also included in the model to account for studies with multiple treatments.
All linear and nonlinear models were fitted and assessed separately for one of 2 categorical classification schemes based on either MP-level (MPA and MPD as described above) or His supplementation method defined as postruminal infusion of free L-His or dele-tion of free L-His from mixture of infused AA (Inf ) and supplementation of His as RPHis, in the form of fat-coated L-His.With the nonlinear model, the model parameters for response variables of DMI, MY, and ECM and marginal recovery could not be estimated on the categorical classification level, and thereby the parameters for these variables are reported for the overall model.Estimation with categorical classification of MPlevel was possible for Eff His for the exponential model, whereas with categorical classification of His supplementation method the comparison was possible for the response MTPY for the log logistic model.Given these fit models, a full dummy variable regression method was then used for each response to compare respective model parameters across classification levels based on single degree of freedom contrasts (Price et al., 2012).For the linear model, interaction of the categorical classification level and factor in the model was tested, and if significant (P < 0.05), included in the model.
All regression analyses were performed using SAS (version 9.4; SAS Institute Inc., Cary, NC).Nonlinear regression used the SAS procedure Proc NLMIXED while linear models were fit using Proc GLIMMIX.

Data Set and Summary Statistics
Characteristics of the publications used in the current analyses are given in Table 1.In 10 studies, His was supplemented to the diet as RPHis and the value reported in each publication for the amount of dHis supplied from RPHis was used in the current analysis.In 1 experiment (Giallongo et al., 2017) 2 basal diets with different dHis levels were used, and this experiment was grouped with the RPHis supplementation method for the purposes of this meta-analysis.In the remaining experiments free His was infused into the abomasum (4 studies), the jugular vein (3 studies) or deleted from a mixture of an abomasally or duodenally infused AA mixture (4 studies).The diets were based on corn silage in 14 and on grass silage in 8 studies.
Summary statistics within each treatment category (control and His supplementation) and His supplementation method (deletion, infusion and RPHis) are given for dietary variables in Table 2 and for animal variables in Table 3.The comparisons based on the summary statistics between control and His-supplemented diets as well as His supplementation methods were made on numerical basis based on calculated mean and standard deviation.Dietary CP and NDF contents were similar for control and His-supplemented diets, but differed among His supplementation methods, both CP and NDF being greatest for infusion, and NDF lowest for RPHis studies.Further, NE L and MP supplies were also similar between control and His-supplemented diets, whereas studies with RPHis had the greatest and infusion studies had the lowest overall supplies of NE L and MP among the 3 His supplementation methods.As expected, the estimated adHis supply was lower for the control than His-supplemented diets, 36.8 g/d versus 49.4 g/d, respectively.Overall, Eff His was greater for the control compared with His-supplemented diets, and greater for deletion and infusion studies compared with RPHis studies.
Experiments investigating His supplementation in lactating dairy cows included in the current analysis represent a wide range of experimental settings in terms of design, basal diet, method and level of His supplementation as well as production level of the animals.All these factors play into the degree of response to His supplementation observed in each experiment and within each His supplementation method.Previous meta-analysis on AA supplementation have indicated, that it can be challenging to test and conclusively include all different factors affecting the production responses to individual AA supplementation (Patton, 2010;Zanton et al., 2014).This is due to complex interactions with dietary factors as well as the complex metabolism of AA (Patton, 2010).Indeed, individual AA are not only used to support protein synthesis but also for other functions including signaling molecules in several hormone responses, gene expressions and other physiological pathways, including gluconeogenesis (Nichols et al., 2019).In addition, and in synergy with the above listed functions, AA also play a role in the utilization, and as a source of energy (Omphalius et al., 2019).
Lastly, studies included in the analysis spanned over several decades and cows in the earlier studies with grass silage-based diets (and infused His) produced on average less milk compared with cows in more recent studies, which were primilary performed using RPHis.The level of dHis in the control diet also affects the response to supplemental His (Lapierre et al., 2021).Therefore, it is important to note that studies using RPHis also had a greater supply of dHis from the basal diet and a greater dose of dHis from supplemental His compared with infusion studies.This may have further affected the level of response to additional His compared with infusion/deletion studies, as discussed below.Overall, however, the data set used for the current meta-analysis provides a robust basis for testing the effect of His supplementation on production parameters and Eff His across different basal diets and methods of His supplementation.

Effects of His Supplementation on Production Parameters
Meta-analysis for the overall effect of His supplementation, reported as SMD, on lactational performance variables are shown in Table 4. Forest plots illustrating the effect of His supplementation in individual studies, and the overall effect size across studies are given for DMI, MY, and MTPY in Figures 1, 2, and 3, respectively.Sensitivity forest plots are given in Supplemental Figures S1, S2, and S3 (https: / / scholarsphere .psu.edu/resources/ c2bfa303 -70bd -4297 -8fb3 -e7399271e938; Räisänen et al., 2023).While these sensitivity analyses indicate some studies affected estimates, the overall assessment shows that only the magnitude of each estimate was reduced and the direction and significance of the final estimates were not affected.
In the full analyses, DMI, MY, MTPY and milk lactose yield were greater (P ≤ 0.01) for cows receiving His supplementation versus control diet.Milk true protein concentration was greater (P = 0.02), and milk fat concentration was lower (P < 0.001) for the Hissupplemented cows.Lastly, plasma His concentration was greater for cows receiving His supplementation (P < 0.001).
The effects of His supplementation with MPA versus MPD diets are shown in Supplemental Table S1 (Räisänen et al., 2023).Similar to the overall effect, His supplementation increased (P ≤ 0.001) MY, MTPY, and plasma His concentration, and decreased (P ≤ 0.03) milk fat concentration for both MP-levels, and the effect size was −0.381 for MPA and −0.470 for MPD diets, respectively.In addition, DMI tended to be greater (P = 0.06) for cows fed MPD diets with His supplementation versus control but no DMI effect was observed for cows on MPA diets.Milk lactose yield was greater (P = 0.01) for cows receiving His supplementation on MPA diets and tended to be greater (P = 0.09) for cows on MPD diets with His supplementation.
The initial meta-analyses performed here supports the conclusions of several previous individual experiments showing the benefit of supplemental His on DMI, MY, and MTPY in lactating dairy cows.The SMDbased effect sizes can be considered moderate for DMI, MY, and MTPY with raw mean differences of 0.228 ± 0.366 kg/d, 1.81 ± 0.661 kg/d and 0.068 ± 0.0469 kg/d, respectively.The main driver for the positive response to His supplementation has been suggested to be its low concentration in microbial protein in relation to its concentration in milk protein (Kim et al., 1999;Patton et al., 2014;Räisänen et al., 2021b).Indeed, a recent meta-analysis by Sok et al. (2017) further confirmed the relatively low His concentration in rumen microbial   protein.This would suggest that cows fed MPD diet, usually relying on a larger proportion of MP from MCP, would be more responsive to His supplementation than cows fed MPA diets.Dry Matter Intake.First, it is noted that DMI was limited to 97% of previous ad libitum intake in Lapierre et al. (2021).However, the study was kept in the meta-analyses because deletion of His decreased DMI, generating refusals, and hence His supply tended to have an effect on DMI in one study of that publication.Overall, DMI response to His supplementation was more pronounced for studies with MPD diets than MPA diets.Amino acids in general, including His, have been suggested to play a role as signaling molecules directly involved in feed intake stimulation, at least in monogastric animal models (see discussion in Giallongo et al., 2015 andRäisänen et al., 2021a,b).Overall, the severity of His deficiency can affect DMI response, and the mechanisms by which His may stimulate feed intake are most likely dependent on the production level of the cow and the overall AA balance in the basal diet.This may have contributed to the numerical difference in the effect size between MPD and MPA diets.Overall, the results from this meta-analysis confirm the positive effect of His on DMI, but the exact mechanisms as to whether the increased intake is a result of a specific signaling pathway of His, or due to a "pull effect" resulting from increased nutrient demand due to enhanced MY (Huhtanen et al., 2011), remains unclear.
Milk Yield.The increase in MY with His supplementation demonstrates the potential of His to be limiting for milk production with diets currently fed to dairy cows.It is important to note that both Lys and Met were supplemented in most of the experiments used in this meta-analysis, and hence the effects of His were not reflective of situations were Lys or Met supplies were limiting.Interestingly, such clear effect on MY has not been demonstrated for Met or Lys supplementation without His, as evident from 2 meta-analysis on Met supplementation (Patton, 2010;Zanton et al., 2014) and experiments supplementing RPMet or RPLys (e.g., Lee et al., 2012;Morris and Kononoff, 2020).Balancing for Met, however, has been proven important for milk protein secretion (Patton, 2010;Zanton et al., 2014), and both Lys and Met are considered to be potentially limiting AA in various types of diets, using different feeding models (e.g., Norfor, 2011;INRA, 2018;NASEM 2021).Interestingly, Lee et al. (2012) demonstrated, that supplementation of all 3 AA (i.e., His, Lys, and Met) restored the MY of cows fed an MPD diet to the MY level of cows receiving an MPA diet (38.5 vs. 38.8 kg/d).In the same study supplementation of Met and Lys only resulted in MY similar to that of MPD (36.9 vs. 35.2kg/d), although not significantly different from the MPA diet.Hence, with a type of rations usually fed to dairy cows, the supply of these 3 AA needs to be carefully considered to optimize the lactational performance of dairy cows.This is especially important in low protein diets as demonstrated by Omphalius et al. (2019), and the current meta-analysis (see below).It has been suggested that the supply of AA being absorbed and available to the mammary gland is important for maximal milk production, and a shortage in one or more AA can become a limiting factor for milk and milk protein secretion (Raggio et al., 2004;Cant et al., 2018).Indeed, in the recent NASEM (2021) model, a multivariate equation, including the supply of 5 individual EAA, among which His, Lys, Met plus Ile and Leu, plus the sum of all other AA is now proposed to predict MTPY.Milk True Protein Yield.In the current analysis, MTPY was the response variable mostly benefiting from His supplementation.In line with our data, this positive response has been especially evident in studies that have supplemented His in the form of abomasal infusion (e.g., Korhonen et al., 2000;Huhtanen et al., 2002;Lapierre et al., 2021; see further discussion below) and when the basal diet was low in RUP, as in grass silage-based (e.g., Vanhatalo et al., 1999;Korhonen et al., 2000) or MPD diets (e.g., Räisänen et al., 2021b).Further, our analysis showed 3.3 and 1.3 times greater milk true protein concentration and MTPY responses, respectively, to His supplementation in studies where MP supply was at least 5% below, compared with diets where MP supply met or exceeded, NRC ( 2001) requirements.This further suggest that cows fed MPD diet can be more responsive to His supplementation than cows fed MPA diets.It should be noted that MP-requirement as described in NRC (2001) assumes a fixed Eff MP although it is recognized that Eff MP is variable and related to energy supply (NASEM, 2021).However, energy supply did not differ between control and supplemented His treatments in the studies included in the current meta-analyses.Therefore, dividing the data into MPA and MPD served to demonstrate the differential response to His supplementation caused by altered AA-profile related to MP supply (i.e., lower His supply) when diets are formulated to be deficient in MP according to NRC (2001), as demonstrated here.
The data from the current meta-analyses, as well as previous animal experiments, establish that a shortage of His supply might be limiting milk protein secretion.Considering all AA in ration formulation will be increasingly important, as the dairy industry aims for optimal utilization of resources for milk production (Schwab and Broderick, 2017).This includes the utilization of the unique ability of ruminants to convert fibrous, human inedible plant mass to high quality milk protein while simultaneously minimizing excess excretion of N into the environment.Hence, low protein diets will be an incremental part of nutritional management of dairy farms to reduce costs, nutrient losses, and emissions (Schwab and Broderick, 2017).Therefore, His, together with Met and Lys, supplementation may be an important element of such feeding regimens to optimize MTPY and MY.However, the optimal supply of His, and other AA, in practical on-farm settings will not only be determined by the lactational performance response to increased AA supply, but obviously also by cost-effectiveness of the supplemental RPAA.(g/d) and conversely increased with increasing Eff His (Supplemental Figure S4; Räisänen et al., 2023) Dry Matter Intake and Milk Yield.A nonlinear response of production parameters to increasing protein or dEAA supply is generally expected based on decreased utilization efficiency with increasing supply (Lee et al., 2015;Omphalius et al., 2019;NASEM, 2021), and was also demonstrated in the current analyses.Both DMI and MY had a lower response at low adHis supply and reached 50% of the maximum response at adHis supply of 45 and 44 g/d, respectively.The relationship between DMI and MY and increasing adHis supply further show the overall positive effect of dHis on the lactational performance of dairy cows, discussed above.Further, the diminishing response in MY at greater adHis supply is reflective of the decreased utilization efficiency of His, as described below.
Milk True Protein.Milk true protein yield also increased with increasing adHis dose, with a greater response in MTPY for Inf compared with RPHis; level of adHis supply to reach 50% of the maximum response was at adHis supply of 39 vs. 50 g/d for Inf vs. RPHis, respectively.Indeed, among the individual studies, Inf studies have consistently shown a positive response in MTPY (e.g., Kim et al., 1999Kim et al., , 2001;;Vanhatalo et al., 1999;Lapierre et al., 2021).The deletion studies especially are effective in inducing His deficiency while simultaneously ensuring a balanced supply of other EAA, and therefore have shown a stronger MTPY response to increasing dHis dose compared with other forms of His supplementation (Lapierre et al., 2021).The difference in MTPY response between His supplementation methods likely stems from multiple factors.First, the difference in the 2 forms of His (infused free His vs. fat-coated His) may affect the rate of absorption.Indeed, it has been shown that intestinal digestibility of Lys increased as the amount of oleic acid in the coating material increased (Wu et al., 2012; i.e., the coating material of RPAA has an effect of its digestion in the small intestine).Another important factor is the estimated bioavailability of the RPHis products used in animal experiments.Based on the difference between the His supplementation methods in the current analysis, dHis supply from RPHis seems to be generally overestimated.As discussed extensively in Räisänen et al. (2020), the estimation of bioavailability of RPAA products is challenging.As follows, the estimated versus true supply of dEAA from these products will vary depending on which method is used to determine their bioavailability and will rarely be completely accurate or comparable to infusion of free His.Further, differences in the types of control treatment between studies can also play into the magnitude of response between His supplementation methods; level of His deficiency was likely greater in infusion and deletion studies (i.e., the response in MTPY) would be greater due to greater Eff His at low basal dHis supply (29.4 ± 1.97 g/d vs. 47.6 ± 9.42 g/d, respectively).Thus, it is likely that   several of these factors contributed to the difference in MTPY response observed between His supplementation methods.
Plasma His Concentrations.As alluded to previously, the magnitude of change in plasma His concentration with increasing dHis dose was also greater for infused His compared with RPHis and was likely caused by similar factors as discussed above for MTPY.With plasma His concentration, the overestimation of bioavailability of RPHis as well as the level of dHis in the basal diet most likely contribute more to this difference than for MTPY.In addition, plasma His concentration is dependent of total dHis supply, as well as the rate of liver extraction and catabolism (Lapierre et al., 2021), and thereby at greater supply, plasma His concentration can reach a plateau where after we detected no further response.Indeed, a dose-response experiment with RPHis on a MPA corn silage-based diet reported no plasma His response to increased doses of RPHis (Räisänen et al., 2021a).In contrast, 2 experiments with MPD diets (i.e., lower dHis level in the basal diet) and similar RPHis product reported a linear increase in plasma His concentration with increasing RPHis dose (Zang et al., 2019;Räisänen et al., 2021b).It was suggested that the lack of response in plasma His concentration resulted from a decreased uptake of additional available His by the mammary gland at higher dHis doses and an increased hepatic removal of His (Räisänen et al., 2021a).The published data with infusion of His, included in the current analysis, have been mainly obtained from cows receiving a grass silage-based diet (Kim et al., 1999(Kim et al., , 2001;;Vanhatalo et al., 1999;Korhonen et al., 2000;Huhtanen et al., 2002), or a low protein diet with infusion of free AA (Lapierre et al., 2021), and thereby the level of dHis from the basal diet was generally low.Further, the infused His doses have been lower, and thereby, together with a low level of His in the basal diet, can lead to greater increments in plasma His concentration in relation to the control treatments in these experiments compared with the plasma His increments observed in experiments with RPHis.
Efficiency of Utilization and Marginal Recovery of His.The negative relationship between Eff His and increasing dHis dose was expected and follows the principle of decreased Eff AA with increasing supply, demonstrated by Giallongo et al. (2015), Omphalius et al. (2019) and NASEM (2021).In addition, the overall Eff His and the rate of change in Eff His were greater for MPD versus MPA diets, indicating a more immediate response to additional dHis supply when MP was limiting.The deciphering of an optimal efficiency of His supplementation, and scoping the optimal dHis supply, has proven challenging.This is partly due to the complex and unclear role of labile His pools, including muscle dipeptides and hemoglobin, in masking the true deficiency of His in lactating cows (Lapierre et al., 2021;Räisänen et al., 2022).Lapierre et al. (2021) suggested an optimal Eff His of 0.77, using predicted NRC (2001) supply, whereas in NASEM (2021) the recommended target Eff His is 0.75.These targets were supported by a dose-response experiment with an MPD diet conducted by Räisänen et al. (2021b).Based on the regression model in the current analyses, adHis supply at an Eff His of 0.77 would be around 60 g/d, for a MTPY of around 1.1 kg/d.This is similar to the suggested optimal His supply and MTPY response by Räisänen et al. (2021a;2022) with MPA diets, but lower than the maximal MTPY response achieved by Räisänen et al. (2021b) with an MPD diet (i.e., Eff His of 0.71 at dHis supply of 74 g/d).In contrast, many of the infusion studies had an Eff His above 0.77, indicating that cows in these experiments were still deficient in dHis even with supplemental His.This may have further driven the difference in MTPY and plasma His response between His supplementation methods, discussed above.
The Eff His was also related to the efficiency of utilization of other EAA, namely Eff Leu , Eff Lys and Eff Met in the current meta-analysis.This indicates an important dependency between individual EAA in terms of their utilization for milk protein secretion, rather than one single AA limiting milk protein synthesis.Based on our calculations and regression analyses, Leu, Lys, and Met are important AA in determining Eff His and should be considered when determining the requirement of His with the types of diets that were fed in the studies included in the current meta-analysis.
Lastly, the average marginal recovery of His across studies was high (19.7% ± 15.68), and we observed an expected decrease in the recovery rate with increasing adHis supply, and an increased marginal recovery with increasing Eff His .The marginal recovery in the current study is greater than for example that reported in Zanton et al. (2014;7% marginal recovery) for supplemental Met.This further suggests His to be an important AA for milk protein synthesis, in addition to at least Lys and Met.
Efficiency of Utilization of His in Relation to Energy Supply.Interestingly, we detected an even stronger relationship between Eff His and the ratio of adjusted dHis to NE L supply and a little spread across studies regardless of His supplementation method, than with adHis supply alone.Further, as above for the relationship between Eff His and adHis supply, the overall Eff His and the rate of change in Eff His were greater for MPD compared with MPA diets.Similarly, in a recent meta-analysis, the ration of MP/NE L or AA/NE L supplies, and their squared term, was the best predictor of the efficiency of utilization of MP or individual EAA (Lapierre et al., 2020).Similarly, in the multivariate equation proposed by NASEM (2021) to predict MTPY, energy supply is the most important term.This demonstrates an important metabolic relation between dietary energy, protein and AA supply.To achieve Eff His of 0.77, dHis to NE L ratio should be around 1.6 g of dHis per Mcal of NE L ; above this ratio, energy rather than dHis would be limiting, and additional dHis from the diet would not be used for milk protein synthesis.
Reversely, at ratios below the 1.6 value, dHis would be limiting milk protein synthesis and the cow would increase MTPY in response to additional His supplementation.Several previous experiments have indicated the effect of not only dietary protein but also energy supply on utilization of AA for milk protein synthesis and yield (e.g., Huhtanen et al., 2002;Lemosquet et al., 2010;Omphalius et al., 2019).Further, Omphalius et al. (2019) demonstrated that increases in both energy and protein supply resulted in positive milk protein yield response, but the mechanisms were different: increased milk protein yield in cows fed high energy diets was due to increased mammary blood flow and subsequent increase in the uptake of NEAA and EAA, whereas cows fed high protein diets had an increased uptake of EAA and decreased uptake of NEAA, without changes in mammary blood flow (Omphalius et al., 2019).In addition, a high energy or high protein diet increased the uptake of group 1 AA, which includes His, and the increment in mammary gland uptake concurred with an increment in milk protein yield.This suggest an increase in mammary transport capacity and intracellular utilization of these AA when dietary energy supply is increased (Omphalius et al., 2019).Hence, dietary energy supply is an important parameter to consider when thriving for a maximal Eff His (or Eff AA ) and establishment of the optimal dietary supply of individual AA.

CONCLUSIONS
Histidine supplementation enhanced DMI, MY, and MTPY, and the positive responses were more pronounced for cows fed MPD diets, confirming the importance of balancing for His for lactating cows fed low protein diets.Regression analyses further showed that DMI, MY, and MTPY increased, and Eff His decreased, in a nonlinear manner, with increasing adHis supply.Furthermore, the magnitude of change in MTPY and plasma His concentration was dependent on the method of His supplementation, being greater for infused His compared with RPHis.This may have resulted from differences in the physical form of His (free His vs. fat-coated His), overestimation of bioavailability of RPHis products, as well as differences in basal diet (grass silage vs. corn silage), level of His in the control treatment and His supplementation level between studies.Further, regression analyses revealed a strong linear relationship between Eff His and the ratio of dHis to NE L supply, highlighting the importance of interaction between energy and AA supply for utilization of His for milk protein synthesis.The difference in response between MP-levels as well as His supplementation methods, and the relationship between dHis and energy supply warrants further research into His requirements with varying dHis and protein and energy supplies, as well as different types of basal diets.Overall, these analyses confirm the importance of His among limiting AA for milk and milk protein synthesis in lactating dairy cows.
of His from a mixture of postruminally infused AA.Infusion = postruminal infusion of His.RPHis = supplementation of His in the form of rumen-protected His. 2 Number of studies.3 Calculated based on NRC (2001), and dietary composition and cow production data if not reported in the publication.4Adjusted net supply of dAA was calculated using the predicted digestive AA flow from NRC (2001) corrected for incomplete recovery with 24-h hydrolysis and excluding the contribution from duodenal flow of endogenous protein flow; includes supplemental AA (postruminally infused AA or RPAA).dAA = digestible AA, RPAA = rumen-protected AA, dHis = digestible His.Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS

Figure 1 .
Figure 1.Forest plot of the effect of His supplementation on DMI in lactating dairy cows based on standard difference in means.The diamond at the bottom indicates the mean effect size, calculated according to a random effects model.The size of the square illustrates the weight of each study relative to the mean effect size.Smaller squares represent less weight.The horizontal bars represent the 95% CI for the study.

Figure 3 .
Figure 3. Forest plot of the effect of His supplementation on milk true protein yield in lactating dairy cows based on standard difference in means.The diamond at the bottom indicates the mean effect size, calculated according to a random effects model.The size of the square illustrates the weight of each study relative to the mean effect size.Smaller squares represent less weight.The horizontal bars represent the 95% CI for the study.Different letters indicate a subset of data within the same publication.
Figure 7. Efficiency of His utilization (Eff His ) observations by MP-level and across (A) increasing adjusted digestible His (adHis) supply or (B) ratio of adjusted digestible His (adHis) to NE L supplies.Metabolizable protein requirements and supply were calculated based on NRC (2001); MP-deficiency (MPD) was defined as MP supplied at or below 95% of NRC (2001) requirements, and MP-adequacy (MPA) above 95% supply of requirements.
Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS

Table 1 .
Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS Characterization of publications used in the meta-analysis

Table 2 .
Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS Summary statistics for dietary variables for the data set used in the meta-analysis

Table 3 .
Summary statistics for animal parameters for the data set used in the meta-analysis Item Sjaunja et al. (1990) of His from a mixture of postruminally infused AA.Infusion = postruminal infusion of His.RPHis = supplementation of His in the form of rumen-protected His. 2 Number of studies.3EstimatedaccordingtoSjaunja et al. (1990)if not reported in the publication.

Table 4 .
Räisänen et al.: HISTIDINE IN LACTATING DAIRY COWS Effect size 1 and heterogeneity for the effect of His supplementation on lactational performance of dairy cows