Growth and body composition of dairy calves fed only milk replacer at 3 intakes

Determination of energy requirements for growth depends on measuring the composition of body weight (BW) gain. Previous studies have shown that the composition of gain can be altered in young dairy calves by composition of the milk replacer diet. Here, our objective was to determine body composition and the composition of empty body gain in young calves fed increasing amounts of a milk replacer containing adequate CP. Male Holstein calves underwent an adjustment period of 14 d after birth in which they were fed whole waste milk at 10% of BW. Calves were then stratified by BW and randomly assigned to either an initial harvest group (n = 11) or to groups fed 1 of 3 milk replacer amounts and harvested after 35 d of growth. All treatments consumed the same milk replacer containing 24.8% CP (dry matter [DM] basis; from all milk proteins) and 18.9% fat, reconstituted to 12.5% solids. Treatments were milk replacer fed at 1.25% of BW (DM basis; n = 6), 1.75% of BW (n = 6), or 2.25% of BW (n = 8), adjusted weekly as calves grew. Calves fed at 1.25% or 1.75% of BW were fed twice daily and those fed 2.25% of BW were fed 3 times daily. No starter was offered. Post harvest, the bodies of calves were separated into 4 fractions: carcass; total viscera minus digesta; head, hide, feet, and tail; and blood. The sum of those 4 fractions was empty BW, which increased linearly as amount of milk replacer increased. Final heart girth and body length, but not withers height, increased linearly as intake increased. Gain: feed increased linearly with increasing milk replacer. Feeding more milk replacer increased the amounts of lean tissue and fat in the body. The percentages of water and protein in the final body decreased linearly, whereas fat percentage and energy content


INTRODUCTION
Accurate determination of the energy requirement for growth in young calves is essential for establishing nutrient requirements and feeding recommendations.Growth requirements for energy depend on the composition of empty body tissue and, particularly, on the fat content which increases with more rapid rates of gain and as the animal ages (NASEM, 2016).The National Research Council (NRC, 2001) established a new approach for specification of nutrient requirements for dairy calves based on limited older data for body composition of young dairy calves that may not reflect composition of Holstein calves of current North American genotypes.Furthermore, calves in those older studies were fed milk replacers based on skim milk protein, rather than on whey proteins used most commonly at present.The NASEM (2021) requirements used data from several more recent experiments to revise standards for growth.
Conventional calf rearing systems relied on restricted feeding of milk or milk replacer to encourage starter intake.Typically, calves consumed a 20% to 22% CP milk replacer at a rate of 8% to 10% of BW (as fed).Calves allowed to suckle their dams typically will nurse between 6 and 10 times daily and consume milk at 16% to 24% of BW (Lineweaver and Hafez, 1967).It has long been known that growth rates of preruminant calves are increased by increasing the amount of milk or milk replacer that is fed.For example, Khouri and Pickering (1968) reported increased feed efficiencies and growth rates as intake increased from 12% of BW to 18% of BW.Rosenberger et al. (2017) reported that growth increased linearly as milk intake increased from 6 to 12 L/d.
The first 2 mo of a calf's life is the most efficient period for increases in BW and stature (Kertz et al., 1998), due to rates of protein deposition and associated water, and therefore this period of rapid growth can be exploited with the use of less restrictive liquid feeding programs.Growth rates and body protein deposition increased linearly as milk replacer intakes were increased (Diaz et al., 2001).Calves in that study were harvested at common weights rather than a fixed time on feed.It is of interest to determine body composition at a common age when evaluating practical feeding programs.Silva et al. (2017) found that growth rates and body protein content increased when increasing amounts of whole milk were fed to Holstein × Gyr calves.
The objective of this experiment was to determine the body composition of dairy calves fed increasing amounts of a milk replacer formulated to be adequate in CP content and then harvested at the same age.Our hypothesis was that calves fed a milk replacer with adequate protein at increasing amounts would grow more rapidly without changes in body composition.

Feeding and Management of Calves
The experimental protocol was approved by the University of Illinois Institutional Animal Care and Use Committee.Calves were part of a larger experiment as described in Bartlett (2001).Portions of the experiment have been published previously (Bartlett et al., 2006).Male Holstein calves were purchased from an Illinois dealer at less than 1 wk of age.Groups of calves were delivered on May 27, July 22, and September 17, 1999, to the University of Illinois Nutrition Field Laboratory.Calves were housed individually in hutches (Calf-tel; Hampel Corp., Germantown, WI), which were placed on 15 to 20 cm of crushed rock.No bedding was used to avoid consumption of organic material that could confound measured nutrient intakes.Upon arrival in the afternoon, calves were weighed and fed 1.8 kg of whole waste milk.The following day, calves were fed whole non-saleable milk at 10% of BW, and calves remained on this diet for 14 d to allow for adjustment.Water was available at all times but no dry feed was offered.Calves were vaccinated with TSV-2 (Pfizer, Exton, PA) and were given 2 mL of BO-SE (Schering-Plough Animal Health, Union, NJ) within 24 h.Approximately 2 wk after arrival calves were vaccinated with 2 mL of Bovishield 4 (Pfizer, Exton, PA).
Following the 14-d adjustment period, calves were stratified from highest to lowest BW and then were randomly assigned to either an initial composition (baseline) group or to one of 12 experimental treatments.Eleven calves were killed for baseline measurements (2, 5, and 4 in groups 1 to 3, respectively) and 74 remained on their respective diets for 5 wk before slaughter.Data from all 11 baseline calves were pooled and used to calculate starting body composition in treatment calves.Three of the 12 treatments were designed to address the current objectives and only data for those treatments are reported here.Investigators were not blinded to treatments.
To determine the effects of increased nutrient intake on growth and body composition, calves were fed a milk replacer at 3 different rates: 1.25%, 1.75%, or 2.25% of BW (DM basis), adjusted weekly (n = 6 per treatment for the 1.25% and 1.75% groups, n = 8 for the 2.25% of BW group).The calves were fed a whey-protein-based milk replacer (24.8% CP, 18.9% fat, 49.2% lactose, 7.2% ash, 5.1 Mcal/kg gross energy), reconstituted to 12.5% solids with warm tap water.Calves fed at 1.25% and 1.75% of BW were fed twice daily, whereas calves fed at 2.25% of BW were fed 3 times daily.Corresponding liquid feeding rates (as fed) were 10%, 14%, and 18% of BW.The milk replacer (formulated and manufactured by Milk Specialties Co., Dundee, IL) was based on whey protein concentrate, dried whey, lard, and tallow.Milk replacer did not contain growth-promoting antibiotics.Milk replacer was sampled weekly and composited by group of calves.Starter feed was not offered to avoid its confounding effect and greater variability associated with starter intake.
Milk replacer intake was measured daily.Water was offered for ad libitum consumption, and intake was recorded daily.Calves were weighed once weekly and the amount of DM offered was adjusted weekly to maintain the desired feeding rates.Calf withers height, body length, and heart girth were measured once weekly.Calves were monitored several times daily and all observations concerning health were recorded.Fecal scores were assigned and recorded once daily using the following system: 1 = dry, hard; 2 = soft, formed; 3 = pudding like; 4 = mix of liquid and some solids; and 5 = liquid.
Blood samples were collected once weekly via jugular venipuncture at approximately 0700 h, which was before Bartlett et al.: BODY COMPOSITION IN MILK-FED DAIRY CALVES the morning feeding.Blood was drawn into 2 10-mL tubes containing heparin and 1 5-mL tube containing EDTA (Vacutainer; Becton Dickinson Vacutainer Systems USA, Rutherford, NJ).Tubes were placed on ice and were centrifuged within 1 h at 1200 × g for 8 min at room temperature.The plasma recovered was frozen (−20°C) until analyses.

Body Composition Procedures
All calves were humanely euthanized at the University of Illinois Meat Science Laboratory using captive bolt stunning followed by exsanguination.Eleven calves were harvested for initial body composition data following the 14-d adjustment period.The remaining calves were harvested 5 wk later.Calves were weighed before harvest, which was approximately 17 h after the last feeding; this weight was considered to represent shrunk BW.Following exsanguination, blood was collected and the weight recorded.The hide and viscera were removed.The body was separated into 3 fractions: head, hide, feet, and tail (HHFT); viscera; and carcass.The gastrointestinal tract (GIT) was removed and weighed.Digesta was removed from the GIT by rinsing the stomach and intestines thoroughly with water.The empty GIT was then re-weighed, and the amount of digesta in the GIT at the time of harvest was determined as the difference in weight between the full GIT and the empty GIT.Individual weights were recorded for the kidneys, liver, and heart; all internal organs were then pooled to form the visceral fraction.The fractions were refrigerated overnight and processed the following day.The HHFT and carcass were ground twice through a whole carcass grinder (model 801 GP15; Autio Co., Inc., Astoria, OR) fitted with a 1.3-cm plate and then were sub-sampled.The visceral fraction was ground twice through a grinder (Butcher Boy model 52HF; Laser Manufacturing, Los Angeles, CA) and then sub-sampled.The sub-samples were frozen and then re-ground twice through the Butcher Boy grinder, using a smaller die (3 mm).Sub-samples were frozen (−20°C) for later analyses, and another sub-sample was lyophilized.
We assumed that the variation in HHFT composition between calves would be insignificant and that diet would have little influence on HHFT composition.The HHFT fraction was extremely difficult to grind; therefore, HHFT composites were made.A HHFT composite was created for each day that calves were slaughtered.All HHFT from calves used to determine initial body composition were composited and 5 HHFT from the treatment calves were composited, thereby creating 3 initial and 3 final HHFT samples.No other fractions were derived from composites.

Analytical Procedures
Milk replacers were analyzed for gross energy (GE) content using an adiabatic bomb calorimeter (Parr Instrument Co., Moline, IL).Metabolizable energy in milk replacers was calculated as 0.91 × measured GE, based on NASEM (2021).The fatty acid content and profile of the milk replacers was determined by GLC of methyl esters formed by acid-catalyzed transesterification (Sukhija and Palmquist, 1988) using a Supelco SP-2380 100-m fused silica capillary column (Supelco, Bellefonte, PA) in a Shimadzu GC-17A gas chromatograph (Shimadzu Scientific Instruments, Inc., Columbia, MD).Nitrogen was determined using the Kjeldahl digestion procedure according to AOAC (1990); N was converted to protein by multiplying by 6.25 for both milk replacers and body tissues.Blood collected during the harvest procedure was not combined with any fraction and samples inadvertently were not retained for analysis.Consequently, concentrations of DM, protein, and energy were determined following the completion of the experiment in whole blood samples obtained from calves fed and managed similarly at the University of Illinois Dairy Unit.Contents of fat (Novakofski et al., 1989), ash (AOAC, 1990), and water (AOAC, 1990) were determined in frozen tissue samples from each body fraction.Energy was measured in freezedried tissue fractions by adiabatic bomb calorimetry (Parr Instrument Co., Moline, IL).
Concentrations of glucose (kit 315-500; Sigma Chemical Co., St. Louis, MO), urea-N (kit 535B, Sigma), total protein (kit 541-2, Sigma), and NEFA (kit 994-75409; Wako Chemical, Richmond, VA) in plasma were determined by standard enzymatic-colorimetric procedures.Concentrations of insulin-like growth factor I (IGF-1) in plasma were determined by radioimmunoassay after removal of binding proteins using acid-ethanol extraction (Sharma et al., 1994).Recombinant human IGF-I and the primary antibody were obtained from GroPep (Adelaide, Australia).Concentrations of insulin in plasma were determined by using a radioimmunoassay kit (Coat-a-Count Insulin kit; Diagnostic Products Inc., Los Angeles, CA) as modified by Studer et al. (1993).

Calculation of Body Composition and Composition of Gain
Initial composition of calves harvested at the end of the experimental period was assumed to be the same as that of the 11 baseline calves at the start of the experimental periods.Empty body weight (EBW) was the sum of weights of carcass, digesta-free viscera, HHFT, and blood.The composition of gain for each calf was calculated as the amounts of water, protein, fat, and ash at harvest minus the calculated amounts of those compo- nents present in each calf at the start of the experiment, as estimated from the average composition of baseline calves applied to the starting BW of each calf.

Statistical Analysis
Calf was the experimental unit.No data were excluded.Data were analyzed using the Mixed procedure in SAS (version 9.4; SAS Institute Inc., Cary, NC).Block (i.e., group) and calf were designated as random effects, whereas feeding rate was a fixed effect.For variables with repeated measures, the model also contained fixed effects of time (as a repeated factor) and treatment × time.Covariance structures considered were compound symmetric, autoregressive order one, and unstructured; the autoregressive order one structure was found to be most appropriate for all variables based on Akaike's Information Criterion.Initial measurements of BW, length, withers height, heart girth, and blood samples during the adaptation period before treatments were assigned were used as covariates when analyzing the respective stature measurements and blood variables.Orthogonal polynomial contrasts were used to estimate the linear and quadratic effects of increasing feeding rate.To determine ME available for growth (MEg), maintenance ME was calculated according to NASEM (2021) and subtracted from total ME intake, expressed per unit of metabolic BW.To calculate the partial efficiency of use of MEg for retained energy (RE), a mixed model regression was conducted without an intercept using Proc Mixed of SAS.Model residuals were inspected for homogeneity and homoscedasticity.Least squares means and standard errors are reported.Significance was declared at P < 0.05.
The average fecal score for the 35-d experiment did not differ among treatments (data not shown).The number of days with fecal score ≥4 increased linearly (P < 0.003) as feeding rate increased.Calves had fecal scores ≥4 on average of 11.5, 13.5, and 22.4 d, respectively.
Final whole body composition of calves was affected significantly by feeding rate (Table 3).Masses of EBW, carcass, HHFT, blood, and total viscera all increased from baseline and increased linearly with increasing feeding rate.The mass of the GIT increased linearly (P < 0.001) with increased feeding rate, but when expressed as a percentage of EBW the GIT did not differ among feeding rates (P = 0.34).Liver mass increased linearly, both as kilograms and as a percentage of EBW (P < 0.001).Masses of heart (P = 0.009) and kidneys (P < 0.001) increased linearly with increasing feeding rate but did not differ when expressed as percentages of EBW.
The final amounts of water (P < 0.001), protein (P < 0.001), fat (P < 0.001), and ash (P < 0.001) increased linearly as feeding rate increased (Table 4).Expressed as percentages, water (P < 0.001) and protein (P < 0.03) decreased, whereas fat increased (P < 0.001) linearly.The percentage of ash in the body was not affected by feeding rate.
Gain in components was affected by feeding rate (Table 5).Gains of water (P < 0.001), protein (P < 0.001), fat (P < 0.001), and ash (P < 0.001) increased linearly as feeding rate increased.Percentage of protein in gain  6).The estimated ME requirement for maintenance (P < 0.001), ME available for gain (P < 0.001), and RE (P < 0.001) increased linearly as feeding rate increased.Energy efficiency expressed as RE: gross energy intake increased linearly (P < 0.001).Regression of RE on MEg intake relative to BW 0.75 (Figure 1) revealed a regression coefficient of 0.546, indicating an efficiency of ME use for RE of 54.6%.The efficiency of convert-ing MEg to body tissue was not affected significantly by feeding rate.
Intake of MP (P < 0.001), MP for maintenance (P < 0.001), and MP for gain (P < 0.001) increased linearly as feeding rate increased.Protein gain as a ratio to CP intake (P = 0.05) or MP intake (P = 0.05) increased quadratically as protein gain increased, with no further increase at 2.25% of BW.
Concentrations of IGF-1 (P < 0.001), insulin (P < 0.001), and glucose (P < 0.001) in plasma increased as feeding rate increased (Table 7).The quadratic effect of increasing feeding rate approached significance (P < 0.09) for IGF-1, indicating that the difference between 1.25% and 1.75% feeding rates was greater than the difference between 1.75% and 2.25% of BW.The concentration of urea-N in plasma decreased linearly (P < 0.001) as feeding rate increased.Although means of urea-N for  1.75% and 2.25% groups were close, the quadratic effect of increasing feeding rate did not reach significance (P = 0.16).Concentrations of total protein and NEFA were not affected by feeding rate.

DISCUSSION
As DMI and intake of nutrients from milk replacer increased, ADG increased linearly.This was reflected in gains of all tissue components and plasma concentrations of glucose, insulin, and IGF-1, which increased linearly with increased intake.The gain to feed ratio increased linearly as feeding rate increased, indicating that calves used their intakes more efficiently for BW gain as intake increased.Gain to feed ratios ranged from 0.40 to 0.72 for calves on our study.Greenwood et al. (1998) noted that lambs given ad libitum access to feed were 17% more efficient at converting feed to gain than restricted fed lambs, in agreement with our results that increasing feeding rate improved gain to feed ratios.Diaz et al. (2001) also reported that gain to feed ratios were greater as milk replacer intake and ADG increased.The increasing feed conversion efficiency obserbed with the increased feed intake is a result of a greater BW gain and the dilution of maintenance requirements.
We fed calves at 2.25% of BW in 3 meals daily compared with 2 meals for the other treatments, as we were concerned about the large volumes fed to the young calves.Calves fed at 2.25% of BW (18% of BW as fed) did not consume all the milk replacer provided to them during the first few days of the experiment.Within 1 wk, however, the calves in this group consumed all of their milk replacer offered.We do not believe the different feeding frequencies affected our results, because  2 L = linear and Q = quadratic effects of increasing feeding rate.n = 6 for 1.25% and 1.75% groups, n = 8 for 2.25% group\.a previous study comparing 3 meals to 2 revealed only minor differences in nutrient use with a slight advantage for those fed twice daily (Grice et al., 2020).
The general health of the calves was good during the experiment.The mean fecal score for the experiment was not different among groups, but the number of days with scores ≥4 increased as intakes increased.However, this increase was not detrimental to growth and feed utilization, as indicated by the increased final weights, ADG, and gain to feed ratio of calves fed at the higher rates of intake.Roy (1980) reported that scours depends more on the load of pathogenic microorganisms in the calf's environment than on nutrient intake.Research has shown that feeding milk to calves close to ad-libitum intake does not cause scouring (Mylrea, 1966;Huber et al., 1984).Thus, the increased fecal scores in our study represented a "loosening" of feces due to the greater intakes of a liquid diet with no fiber, but not a pathogenic situation.
Fat content of the whole body increased as feeding rate increased but was low for all treatments, in agreement with Diaz et al. (2001).The percentages of protein and water decreased as feeding rate increased, supporting the inverse relationship between fat and water in the body (Reid et al., 1955).In contrast, Donnelly and Hutton (1976a,b) reported only an increase in percentage of fat    in the final composition of the whole body.Although inadequate dietary protein increases the fat content of the body (Bartlett et al., 2006), Diaz et al. (2001) fed a 30% protein milk replacer, presumed to be adequate for all rates of gain, and still observed increased body fat content with increasing feeding rate.Increasing the feeding rate modestly decreased the percentage protein in gain but increased percentage of fat.Diaz et al. (2001) reported that calves fed the lowest amount of DM had the greatest CP in BW gain, similar to our results.Calves fed more DM and gaining more rapidly had greater fat content of BW gain (Diaz et al., 2001), similar to our study.Increasing milk intake resulted in decreasing protein and increasing fat in bodies of Holstein × Gyr calves (Silva et al., 2017).Nevertheless, increasing nutrient intake will not cause disproportionate increases in fat deposition if dietary CP is adequate (Bartlett et al., 2006).
Energy utilization expressed as RE: intake energy increased linearly as intake increased, due to the dilution of maintenance concept.As expected RE:ME intake also increased linearly, but, after subtracting ME for maintenance, there was no difference among treatments for RE: MEg.Regression of RE on MEg intake resulted in a calculated partial efficiency of ME use of 54.6%, which is similar to the value (55%) used by Van Amburgh et al.Protein efficiency, expressed as protein gain divided by either CP or MP intakes, increased quadratically as intake increased.The lower efficiency for calves fed 1.25% of BW resulted from dietary CP supply being greater than the ME supply for its use.Increasing intake to either 1.75% or 2.25% resulted in similar protein efficiencies, indicating that protein and energy supplies were reasonably in balance.The concentration of urea-N in the plasma declined as feeding rate increased, indicating that calves fed at higher rates of intake used protein more efficiently.This agrees with the greater ratios of retained protein to MP intake as feed intake increased.The CP content of the milk replacer was higher than needed for the calves fed the milk replacer at 1.25% of BW (NASEM, 2021), resulting in less efficient use of dietary CP and elevating urea-N.
The IGF-1 and insulin concentrations in plasma were increased as feeding rate increased; IGF-1 was highly correlated (r = 0.72) with BW gains.We infer that increasing nutrient intake increases growth rates at least in part by increasing IGF-1 and insulin concentrations in the plasma, perhaps via the mTOR mechanism (Wackerhage and Ratkevisius, 2008).Bartlett et al. (2006) found that both increasing energy and increasing CP of the diet increased IGF-1.Haisan et al. (2018) reported that IGF-1 was increased in calves fed a high plane of nutrition compared with a low plane of nutrition.Because IGF-1 plays a direct role in integrating the growth, maintenance, repair, and function of the immune system (Clark, 1997), increases in IGF-1 resulting from improved nutrition might be expected to improve the immune status of calves, but this remains to be demonstrated experimentally.

CONCLUSIONS
Our results indicate that feeding rate, which determines energy intake, determines growth performance and efficiency in young calves.Calves fed at a typical rate of 1.25% of BW are not provided adequate energy for maximal growth; this feeding rate actually limits their growth.Feeding calves on a scheme in which they would consume milk replacer with adequate protein at or more than 1.75% of their BW increases the amount of lean tissue and fat deposition and the efficiency of gain.As gain increases, the percentage of protein in gain decreases and the percentage of fat increases.Efficiencies of energy use and protein use increase as feeding rate increases.Increasing IGF-1 and insulin might be key mediators of the effects of increasing milk replacer intake on growth.
Bartlett et al.: BODY COMPOSITION IN MILK-FED DAIRY CALVES Bartlett et al.: BODY COMPOSITION IN MILK-FED DAIRY CALVES Bartlett et al.: BODY COMPOSITION IN MILK-FED DAIRY CALVES Bartlett et al.: BODY COMPOSITION IN MILK-FED DAIRY CALVES

Figure 1 .
Figure 1.No-intercept regression of retained energy (RE) on ME intake for growth for calves fed milk replacer at increasing rates.Y = 0.546X ± 0.022.
(2019)  andNASEM (2021).Perhaps this is not surprising since the data herein were used by both Van Amburgh et al. (2019) and NASEM (2021) in calculation of the energy requirements for young calves.

Table 1 .
Bartlett et al.:BODY COMPOSITION IN MILK-FED DAIRY CALVES Daily intakes (least squares means ± SE) of DM, CP, lactose, total fatty acids, gross energy, and metabolizable energy (ME) for calves fed milk replacer at increasing rates

Table 2 .
Initial and final BW, average daily gains (ADG), final stature measurements, and gain to feed ratio of calves fed milk replacer at increasing rates (least squares means ± SE)

Table 3 .
Final shrunk body weight (SBW), empty BW (EBW), and weights of body components and internal organs for calves fed milk replacer at increasing rates

Table 4 .
Empty body composition of baseline calves and calves fed milk replacer at increasing rates

Table 5 .
Gains in empty body weight (EBW) and empty body chemical components (least squares means ± SE) of calves fed milk replacer at increasing rates

Table 7 .
Concentrations of IGF-1, insulin, and metabolites in plasma (least squares means ± SE) of calves fed milk replacer at increasing rates