Introduction

Reducing feed costs of the dairy calves is one of the critical issues that dairy farmers are faced with. Usually, protein accounts for the highest amount of the ration cost, while it is regarded as an important part of balancing di- ets for livestock. The amount of protein is very important in the diet of pre-ruminant calves, as reduced animal production, welfare, and polluted environment are the consequences of too low or too high protein concentrations. A concern of overfeeding protein is the amount of N that enters the environment; therefore, researchers have conducted research to re- duce the amount of N lost to the environment while optimizing N usage within the cow (Er- ickson and Kalsheue, 2020). However, there are not enough studies on the most suitable amount of crude protein in the solid feed of pre-ruminant calves. As well, although the NRC (2001) has recommended the require- ments of non-synthesizable AA (Amino Acid) for various animal species, their diets remain suboptimal regarding growth, development, reproduction, lactation and health (Guoyao et al., 2014). The NRC model indicates that 18% CP (DM basis) in starter should be adequate for calves fed with enhanced milk replacer, and that energy supply rather than CP supply would be a limiting factor for growth. How- ever, data needed regarding the determination of the necessity to an increased CP or metab- olizable protein concentration in the starter is inconclusive (Stamey et al., 2012).

In dairy cows, the advantage of improving the balance of absorbable AA is the increased efficiency of the use of absorbed AA for milk protein production. It has been demonstrated that improved lysine and methionine nutrition reduced the amount of dietary CP needed to achieve similar yields of milk protein (Socha, 2005). Individual infusion of lysine and me-

thionine post-ruminally, has revealed that these two amino acids are the first and second limit- ing amino acids in dairy cows. The most effec- tive way to increase the availability of methi- onine post rumen is the use of rumen-protected methionine (Rulquin, 2006). Awawdeh (2016), using rumen-protected methionine plus rumen- protected lysine reported improved milk yield and protein contents of dairy cows and the re- sults were better than supplying rumen-protect- ed methionine alone.

In dairy calves, the provision of adequate nutrients from liquid and solid feeds and maintaining average daily gain above  0.5  kg/d can enhance the first lactation perfor- mance of heifers when combined with prop- er post-weaning practices (Gelsinger et al., 2016). Increased Average Daily Gain (ADG) resulted from increasing amounts of CP and ratios of CP to energy in the feeding of calves only fed with milk has been reported in the study by Chapman et al. (2017). Hill et al. (2006b and 2007c) showed that in calves fed with starter, feeding a 26% CP and 17% fat milk replacer supported more ADG than a milk replacer containing 20% CP and  20% fat. Feeding a 20% CP milk replacer at 0.45 kg/d, supplemented by lysine and methionine improved ADG in all whey CP formulas. A 17% improvement was observed in ADG resulted from adding lysine and methionine, maximized with 2.34% lysine. The linear and quadratic ADG response to added methionine was equal to 13 and 7%, respectively, with a plateau appearing at 0.72% methionine. There was no ADG or efficiency response  to add- ed threonine (Hill et al., 2007b).  Stamey  et al. (2012) studied the growth performance of Holstein dairy calves that received different amounts of CP in their solid feed. In terms of solid feed intake and growth performance of

calves, especially around weaning,  moder-  ate advantages were observed in starter with 25.5% CP (DM basis) compared to a starter containing 19.6% CP (Stamey, et al., 2012).

Besides, some studies have reported the effects of amino acids on immunity. Much non-essential AA and essential AA participate in cell signaling, gene expression, and met- abolic regulation. Thus, functions of AA be- yond protein synthesis must be considered in dietary formulations to improve the efficiency of nutrient use, growth, development, repro- duction, lactation, and well-being in animals (Guoyao, et al., 2014).  Amino  acids  might be a tool to transform animals from a pro- to an anti-inflammatory phenotype through the down-regulation of several genes related to immunity system (TLR-4, NF-κB, TNFa, IL- 1β, IL-2, IL-6, IL-8, CCL-5 and CXCL-16)

whose expression increases during inflamma- tion (Tsiplakou, et al., 2018). In a study done by Montano et al. (2016) on feedlot calves, it was found that methionine and lysine were closely co-limiting amino acids.

Greater heifer growth rates can theoretical- ly lead to an earlier breeding event and reduce the time that the heifer spends in a non-pro- ductive state (Roche et al., 2015).

In this respect, one of the strategies that can be used is to increase the metabolizable protein (MP) availability, which can be understood as an amino acids mixture that is absorbed in the small gut, and that can be used for muscle and mammary gland development, among other functions (Cervieri et al., 2001). Once the re- quirements of the RDP have been met, supply- ing rumen undegradable protein (RUP) could improve the supply of metabolizable protein and decrease the portion of nitrogen com- pounds that is recycled to the rumen thereby increasing the availability of nitrogen for ana- bolic purposes (Rufino et al., 2016).

In a study done by Silva (2017) in prepu- bertal and pubertal heifers, it was concluded that Microbial crude protein (CPmic) synthe- sis, microbial efficiency, and the efficient use of N for microbial synthesis decrease as the supplied RUP increases. The intestinal digest- ibility of CP is negatively affected by RUP levels, which is due to the reduction in the CP- mic flow, which probably has greater intesti- nal digestibility than the protein of feedstuffs. The urinary N excretion decrease  according to the RUP supply increase, which is due to the decrease in ruminal N losses as ammonia and the increase in N recycling.

Although categorizing dietary protein into RDP and RUP has not been extensively used for dairy calves, there are studies where adding protein as RUP has produced positive growth response, no response or a negative response (Kazemi-Bonchenari, 2016). In a study done by Kazemi-Bonchenari et al. (2016), the con- centration of alanine aminotransferase was also lower (P<0.05) for calves fed high-fat starter compared to those fed low-fat starter on day 65, and these levels tended to increase with the addition of RUP. In conclusion, no effects were attributable to feeding a high-RUP starter.

Studies that evaluate whether reducing CP level in the starter and adding AA would im- prove dairy calves’ performance or not, were not available in reviewed articles. Thus, the present study was aimed to assess  the effect of crude protein level in the starter and adding rumen- protected and unprotected lysine and methionine on the growth performance, feed efficiency and well-being of calves as well as some blood metabolites.

Materials and Methods

Animals, management, and experimental design

The study was carried out in a commercial

dairy farm (Goldasht Nemouneh) from De- cember 2017 until March 2018, in Isfahan, Isfahan province, Iran.

Forty-eight 3-day-old Holstein calves (38.56 ± 1.75 kg of BW; mean ± SD) were weighted and moved to individual pens (1.2

× 2.5 meter) with a sand bed, replaced every 24 to 48 h as needed. They had been fed with four liters of colostrum within 6 h of birth and colostrum feeding continued for the first 3 days of their lives. Blood samples were tak- en from the jugular vein by venipuncture at 24 h after the first colostrum intake and were sent to the laboratory for analysis. Serum total protein was determined as an indicator of passive transfer of immunity. Neonatal calves with the serum level of protein greater than 6 gr/dl remained in the study (Mojahedi et al., 2018). Calves were bucket-fed twice a day with pasteurized whole milk; all calves were weaned at day 56 of the experiment.

Calves (n=12, six males and six females) were randomly allocated to four treatments in a completely randomized design. Treat- ments were: (1) 18% CP starter without AA (Amino Acid), (2) 18% CP with 0.0340% protected lysine and 0.016% protected me- thionine, (3) 18% CP starter with 0.215% un-protected lysine and 0.012% un-protect-

ed methionine, (4) 22% CP without AA.

Amino acids were added to experimental diets at the feed factory of the dairy farm and mixed thoroughly with other ingredients.  The characteristics and amount of used ami- no acids were as shown below:

(1)  Rumen-protected methionine, Smarta- mineM brand name, manufactured by Adis- seo Co. with 74% purity at 0.016% of the diet.

(2)  DL Methionine, Rhodiment brand name, manufactured by Adisseo Co. with 99% puri- ty at 0.012% of the diet. (3) Rumen protected lysine, LysiPearl brand name, manufactured by Kemin Co. with 50% purity at 0.340% of the diet. (4) L-Lysine, ADM,  manufactured by Archer Daniels Midland Company with 78.8% purity at 0.215% of the diet.

Components of the experimental diets and their nutrients are shown in Table 1. Before the experiment initiation, basal diet (diet  one) was made in the feed plant of the dairy farm, then a sample was taken and sent to the specialized laboratory to determine the cur- rent level of methionine and lysine in the diet using Pico-Tag, HPLC method (White et al., 1986) an additional level of methionine and lysine (20% more) were included in the diets according to the laboratory results.

Table 1. Ingredients, energy and nutrient composition of experimental diets fed to dairy calves for 70 days

Experimental Diets

Experimental Diets

AA= Amino acid; RDP= rumen degradable protein; RUP= rumen un-degradable protein; ME = metabolizable energy; NEG= net energy for growth; NEM= net energy for maintenance

1 Contained stearic acid, 5% of DM; palmitic acid, 30% of DM; oleic acid, 11% of DM; linoleic acid, 50% of DM; and linolenic acid, 4% of DM. 2 Contained (mg/kg) Co, 100; Cu, 4,290; I, 200; Mn, 10,000; Zn, 20,000; Mg, 67,500; and Ca, 240,000.

3 Contained (IU/kg) vitamin A, 1,300,000; vitamin D, 360,000; and vitamin E, 12,000.

4 Rumen protected methionine, SmartamineM brand name, manufactured by Adisseo co. with 74% purity. 5 DL Methionine, Rhodiment brand name, manufactured by Adisseo Co. with 99% purity.

6 Rumen protected lysine, LysiPearl brand name, manufactured by Kemin Co. with 50% purity. 7 L-Lysine, ADM, manufactured by Archer Daniels Midland Company with 78.8% purity.

* Despite adding amino acids to rations in protected or non-protected form, the CP percentage of diets are nearly similar to each other because the percentage of corn is reduced to maintain the total amount of ingredients constant.

Measurements, sampling, and analyses The equal amount of milk was fed to all calves. Starter intake was measured every morning. Calves were weighed by 7-day intervals using an electronic balance. Av- erage daily weight gain and feed efficiency ratio (kg of BW gain/kg of total dry matter intake) were computed for pre-weaning,

post-weaning, and the overall experimental period. Blood samples were obtained from the jugular vein by venipuncture at day 30 and 70 of the study into an evacuated tube containing clot activator 3 h after morning feeding, and they were immediately placed on ice. Tubes were centrifuged at 3,000 × g for 15 min to separate the serum, reserved at

−20 °C for further analysis. Serum creatinine and urea were specified by colorimetric and enzymatic assays (Pars Azmoon Co., Tehran, Iran). Serum IgG was determined according to the immunoturbidimetric method using a commercial kit (Pars Azmoon Co., Tehran, Iran). Total protein was measured photomet- rically based on the Biuret method using a commercial kit (Pars Azmoon Co., Tehran, Iran). Feed samples were taken weekly from all diets, and were oven-dried and kept for subsequent analysis for CP. Skeletal growth including the length of body, body girth, height of withers, height of hip and the width of the hip of the calves were recorded at the start, weaning (day 56) and at the end of the study (day 70) according to the manner char- acterized in the study by Khan et al. (2007). As calves were fed with whole pasteurized milk produced in a dairy farm, two bulk  milk samples from two consecutive days were collected bi-weekly and kept  at 4 °C in tubes containing potassium dichromate as a preservative until analysis by Milkoscan.  A total solid concentration of milk samples was used for the calculation of total whole milk DMI (Dry Matter Intake). Spot urinary samples were taken directly by stimulating the urinary canal on the last day of the ex- periment. These samples were immediately frozen at −15 °C for subsequent analysis.

Statistical analysis

The study was conducted based on a com- pletely randomized design. Data on body weight, starter consumption, ADG, and Gain-to feed ratio (G: F) were analyzed using the MIXED procedure in SAS software (ver- sion 9.4, SAS Institute Inc., Cary, NC) along with Repeated Measures for pre-weaning (from day 1 to 56 of the study), post-weaning (from day 57 to 70 of the study), and over- all period (from day 1 to 70). Initial values

of body measurements were considered as a covariate for the analysis of body measure- ments. Significance was declared at P<0.05 and trends were considered when 0.05 < P

< 0.10. Means analysis was conducted using LSD for the probability. The statistical mod- el was as below:

Yij=µ+ Si+ Ti+ β(Xi-X) + eij

in which Yij is the dependent variable; µ is the overall mean; Si is the sex of calf; Ti is the effect of treatment; and eij is the residual error. The comparison among treatments was done using 3 independent contrasts. The first con- trast compares the level of protein (18% versus 22%) in starter diet; second contrast compares using AA in treatments against non-using AA diets and the third contrast compares protected

AA versus unprotected one.

Results

Feed intake and growth

Starter intake, ADG, and feed efficiency data are reported in Table 2. Calves fed start- er diet containing 22% CP without supple- mented AA had the greatest amount of feed intake, total and pre-weaning ADG, weaning and final weight (P<0.05). However, feed ef- ficiency was not different among treatments (P>0.05), neither during the whole exper- iment nor before and after weaning. Also, starter intake before weaning and ADG after weaning was not different among treatments (P>0.05). The contrasts show the difference between two
treatments (CP18% versus 22%) as calves fed with 22% CP in starter diet had higher feed intake, ADG pre-wean- ing and during whole experiment and body weight (P<0.05). However, there was not any difference between using or not using AA and also between PAA or unprotected one in starter intake, ADG, gain to feed ratio and body weight.

  Table 2. . Effects of crude protein level and adding rumen protected lysine and methionine or non-protected one on intake, ADG, feed efficiency, and BW of dairy calves

AA=amino acid

1 milk consumption is included 2 only from starter

  a,b Values within a row with different superscripts differ significantly at P<0.05.

Skeletal growth

Table 3 shows the data related to skele- tal growth. Body length at weaning and the end of the experiment showed significant differences between treatments as calves received 18% CP with unprotected AA had the highest value (P<0.05). As other skel- etal growth factors did not reveal the sig- nificant difference, it cannot be concluded whether CP level or using protected AA affect skeletal growth. The contrasts show that only protected AA versus unprotected

one, resulted in a difference in body length (P<0.001).

Immunity measurement

Immunity measurements of calves includ- ing IgG and total protein concentration in se- rum are presented in Table 4. IgG and total protein in serum did not show any significant differences between treatment neither at day 30 nor at day 70 (P>0.05). Also there was not any difference between two levels of CP in the starter diet, using AA or not using it and the kind of AA (P>0.05).

                     
Table 3. Effects of crude protein level and adding rumen protected lysine and me- thionine or non-protected one on skeletal growth of dairy calves

Item

Experimental Diets

SEM      P-Value

18CP vs

P-Value 18CP vs

PAA vs

AA= amino acid

a,b Values within a row with different superscripts are significantly different (P < 0.05)

Table 4. Effects of crude protein level and adding rumen-protected lysine and methi- onine or non-protected one on immunity measurement of dairy calves

AA= amino acid; IgG =immunoglobulin G; d=day

a, b Values within a row with different superscripts are significantly different (P<0.05)

Health score

The results of health scores are pre- sented in Table 5. Health  scores  except  for fecal score, was not different between treatments. Calves received 18% CP with unprotected AA had the lowest (better) fecal score (P<0.05). Contrasts between treatment groups showed that using AA versus not using it and protected AA ver- sus unprotected one resulted in a lower fe- cal score (P<0.05).

Urea and creatinine of serum

Results of serum urea and creatinine are presented in Table 6. There were no differ- ences in serum creatinine and urea between treatments on day 30 of the experiment. However, the dietary effect on urea was sig- nificant at day 70 (P<0.05). Contrasts show that calves that received 18% CP, had less serum urea at day 70 (P<0.05), and calves fed with 22% CP in starter diet had less se- rum creatinine at day 30 (P<0.05).

Table 5. Effects of crude protein level and adding rumen-protected lysine and me- thionine or non-protected one on health score of dairy calves

a, b Values within a row with different superscripts are significantly different (P < 0.05)

Serum amino acids level

The amount of lysine and methionine amino acids in serum are presented in Table

1. There was not any significant difference

among treatments (P>0.05). Also, there was not any difference between the level of CP, using AA versus not using it, or the kind of AA (P>0.05).

Table 6. Effects of crude protein level and adding rumen-protected lysine and me- thionine or non-protected one on serum parameters of dairy calves

AA= amino acid; IgG =immunoglobulin G; d=day

a, b Values within a row with different superscripts are significantly different (P<0.05)

Discussion

In the study done by Lee et al. (2012), cows received lower CP in their diet and had lower feed intake too, which is in agreement with the present study. Also, in their study, adding protected lysine and methionine led to the same amount of feed intake that is again in line with the present study as there was no difference between treatments with the addi- tion of AA, either protected or unprotected. So, maybe using AA in any form does not have any effect on calves feed consumption. In the study done by Tamura et al. (2019) no differences were found in milk yield, milk composition or cow health between cows re- ceived protected methionine and those who did not receive. Also, in the present study, there was no significant difference in perfor- mance (gain to feed ratio) of dairy calves.

In another study conducted by Socha et al.

(2005) on cows in a transition period, there was no difference in feed intake with 2 levels of CP (16 and 18.5%) and basal diet and diet supplemented with rumen- protected lysine and methionine. While in the present study, during the whole experiment, higher CP in the starter led to higher feed intake which is probably due to higher total absorbable pro- tein in diets with higher CP level.

In a study done by Kazemi-Bonchenari et al. (2016) on pre-weaned calves, there was no interaction between RUP and fat, nor  was there any effect of high RUP on growth characteristics that are in line with the pres- ent study. It has been found that  greater RUP content improved feed efficiency in dairy calves by reducing the intake but not increasing the gain (Kazemi-Bon-chenari et al., 2015). In the present study, no beneficial effects of high RUP content were observed

related to ADG or feed efficiency which was in agreement with Swartz et al. (1991). Be- cause the rumen is not fully functional in pre- weaned calves, providing RUP with protect- ed amino acids may have had little effect on the amounts of AA composition of protein in the small intestine. More research with other sources of RUP might be able to determine if delivery to the small intestine of a RUP with different AA compositions may increase calf performance. Incomplete development of ru- men function as well as an immature resident microbial population in pre-weaned calves, which may result in  similar  RUP contents of feedstuffs (Holtshausen and Crywagen, 2000), may explain why there are no con- clusive benefits to formulating starters with higher RUP sources in the present study.

Variation in the starter intake by calves is the cause of more than half of the variation in weight gain in the same period. But in the present study, although calves that consumed a higher amount of starter had higher ADG, there was not a difference in feed efficiency between treatments. So, neither CP level nor the kind of AA affected these parameters. Although 18% of CP provided more micro- bial protein, 22% CP level provided more absorbable protein which neutralized their effect on each other, and led to a lack of feed efficiency between treatments.

The observed differences between skeletal growth factors were the result of low repeat per treatment or management. In a study done by Stamey et al. (2012), the kind of starter (two-level of CP; enriched and convention- al) did not affect on skeletal growth which  is in agreement with the present study.  But in a study done by Tahmasebi et al. (2014), the protein source of the starter affected heart girth size that is not in line with this study.  In a study done by Margerison, et al. (2013),

at weaning, calves fed whole milk plus ami- no acids and plant carbohydrates had great- er mean BW gain, a lower number  of days to target BW, and a greater mean hip-width gain compared with calves fed with whole milk, which is not in agreement with present study as there were no differences between treatments that received amino acids and treatments that did not. But in their study, the mean gain in hip height did not differ among treatments, which is line with our study.

Moallem et al. (2004) found that increasing RUP in post-weaning heifers  could  be  used to accelerate simultaneous increases in skel- etal growth rates, whereas Kazemi-Bonche- nari et al. (2015) observed no effect of RUP level on body measurements in pre-weaning calves. Bethard et al. (1997) indicated that in post-weaning heifers, the RUP level had no ef- fect on hip height but that dietary energy great- ly increased it. Generally, it appears that in the pre-weaning starter, the RUP level does not in- fluence the body measurements of calves.

Total serum protein is  correlated  with IgG, and total protein above 5.2 gr/dl indi- cates the good inactive immunity transfer in a healthy calf (Wilm, 2018). As contrasts in Table 4 show, neither CP level nor the kind of AA had any effect on IgG concentration and total serum protein, indicating that di- etary treatments did not have any effect on the immunity of calves. In the study con- ducted by Vailati-Riboni et al. (2017), using methionine pre-calving led to a higher im- mune response which is not in agreement with the present study. In a study done by Senevirathne et al. (2017), the level of CP did not affect diarrhea frequency which is in agreement with the present study as a fecal score of calves that consumed 22% CP was not different with calves that consumed 18% CP. The results of the study by Brscic et al.

(2014) showed that cows that received more CP/d, had fewer days of treatment for respi- ratory disease. While in the present study, the respiratory score was not different among treatments (P>0.05) that is in agreement with a study done by Ghassemi et al. (2013) that showed that different level of starter intake (and consequently different amount of CP), had no significant effect on respiratory score and days of drug administration for pneumo- nia and diarrhea. Besides, in a study done by Silva et al. (2018), supplementation of milk replacer with lysine and methionine and the association with glutamate and glutamine  did not affect performance, fecal scores or metabolism of calves.

So it appears that neither CP level nor the kind of AA has any effect on the immunity and health situation of calves. That maybe because their basal diet was not deficient in essential amino acids.

In a study done by Klemesrud et al. (1998) adding rumen-protected lysine and methi- onine to the diet of fattening calves, did not have any effect on protein utilization which is in agreement with present study as there was not any difference between rumen-pro- tected and unprotected ones.

Blood urea nitrogen concentration has a positive linear relationship with dietary CP intake, its ruminal degradability, and resul- tant ruminal ammonia concentration in cattle (Lohakare et al., 2006). More protein intake because of greater solid feed consumption and its ruminal degradation has probably re- sulted in greater concentrations of BUN in calves fed diet with higher CP. Greater con- centration of BUN is also an index of renal dysfunction (Khan et al., 2007b);  however, in this study the blood creatinine concentra- tion in all calves was in the safe range and did not differ between treatments.

In a study done by Lee et al. (2012b), supplementing diets that were low in con- centration of microbial protein with ru- men-protected methionine and lysine, did  not increase the serum level of lysine or me- thionine which is in agreement with the pres- ent study. In another study done by Rulquin et al. (2006), serum methionine concentra- tion increased 110 and 65 percent after add- ing methioninand Smartamine respectively. In the study done by Tsukano et al. (2017), results showed that increases in plasma total amino acid (TAA) and branched-chain ami- no acid (BCAA) concentrations in diarrheic calves with severe acidemia were the result of an acceleration in proteolysis, similar to that in humans. In a study done by Tsukano and Suzuki (2019), concentrations of plasma essential amino acids, non-essential amino acids, branched-chain amino acids, gluco- genic amino acids, and ketogenic amino ac- ids in diarrheic calves with hypoaminoaci- demia were significantly  lower  than  those in healthy calves. No significant differences were observed between diarrheic calves with normoaminoacidemia and healthy calves when looking at these parameters that are in line with the resent study, because the con- centration of serum lysine and methionine in calves with a low or high fecal score, was not different as they all were normoaminoac- idemia. Lack of increase in methionine may indicate that diets were not deficient in ener- gy or amino acids as a shortage of energy or other amino acids probably responds to me- thionine supplementation.

In conclusion, the results of the present study showed that higher CP levels in calf starter and adding amino acids to starter diet, either in protected form or unprotected one, led to some significant differences in ADG and body weight but did not result in overall