Male Holstein calves were managed according to the recommendations of the Animal Care Committee of Institut de Recerca i Tecnologia Agroalimentàries (IRTA) under the approval research protocol FUE-2017-00587321, authorization code 9733. Holstein male calves were obtained from Granja Murucuc (Gurb, Barcelona, Spain) and housed individually at IRTA in Torre Marimon (Barcelona, Spain). Plastic ear tag identification with the animal's number was used. In all cases, animals were under veterinary control and no health problems were observed during the experimental period. The total number of individuals included in the study was n = 193. Animals belonged to three trials.

In Trial 1 (T1), 60 new-born calves (age = 0 h) were fed with colostrum within the first 6 h after birth. At 24 h of age, blood samples were taken. Then, at age of 3 ± 1.3 days old, animals were fed with a MR and blood samples were collected at 2 and 5 weeks of age 4 h after the morning feeding. The T1 was carried out between April and July 2017.

In Trial 2 (T2), 76 calves (average age = 3 ± 1.7 d) were fed with MR and blood samples were collected at 2, 5, and 7 weeks of age 4 h after the morning feeding. The T2 started in December 2016 and ended in March 2017.

In Trial 3 (T3), 57 calves (average age = 5 ± 2.2 d) were fed with MR and blood samples were taken at 2, 5, and 7 weeks of age 4 h after the morning feeding. The T3 was carried out between April and August 2018.

All animals followed the same MR feeding program that consisted on feeding 2-L of MR (Nukamel, Weert, The Netherlands) twice a day containing at 12.5% concentration of MR powder (500 g/d) for the first 4 days, subsequently MR was increased to 5 L/d of MR at 12.5% concentration of MR powder (625 g/d) for the next 3 days, then MR was increased to 6 L/d of MR at 12.5% concentration of MR powder (750g/d) for the next 7 days and then 2 meals of 3 L at 15% concentration (900 g/d) until 49 d of age. Then, MR was limited to a single offer of 3 L also at 15% (450 g/d) until end of the study. Animals also had access to water and chopped barley straw ad libitum. Chemical composition and ingredient of the MRs and the starter intake information are shown in Tables S1, S2.

A total of 336 commercial crossing piglets ([Large White x Landrace] x Pietrain), male and female 21-d old piglets of 4–7 kg of body weight (BW) were weaned. The animals were moved from the farrowing to the nursery unit (within the same commercial farm without transport) in the morning on the day of weaning. Plastic ear tag identification with the animal's number was used. In the nursery room, pigs were then allocated in 24 pens (14 piglets/pen). Each pen has had a commercial non-lidded hopper and a nipple waterer to ensure ad libitum feeding and free water access. The experimental procedure has been approved by ethical committee (CEEAH) of Universitat Autònoma de Barcelona (Registration n°:1406 of the Departament de Medi Ambient i Habitatge of the Generalitat de Catalunya).

At weaning, animals were offered the same basal pre-starter diets following the same specification. The pre-starter and starter diets were fed ad libitum for fourteen consecutive days (P1) and 21 consecutive days (P2), respectively. A basal diet was formulated to contain 2,470 kcal NE/kg; 20.5% CP/kg and 1.35% Dig Lys for P1 and 2,400 kcal NE/kg; 18.1% CP/kg and 1.20% Dig Lys for P2. Diets were formulated to meet the requirements for growth of newly weaned piglets (15). No antibiotics and ZnO were included in the basal diets.

Blood samples from 3 piglets per pen (a total of 72 animals) were collected avoiding cross-contamination between pens at 8 am at week 3 of age (day 0 after weaning), week 4 (day 7 after weaning), and week 5 (day 14 after weaning). Piglets were individually weighted before weaning and distributed for a balanced body weight within each pen. Piglets were sorted by descending BW within each pen and the one in the middle (median), the ones immediately above and below the median were picked out to be sampled. The same animals per pen were sampled on weeks 4 and 5, respectively.

Blood samples obtained from jugular venipuncture were collected in 10 ml Vacutainer tubes without anticoagulant. Sera were obtained by centrifugation at 1,500 g for 10 min and transferred to new tubes to be stored in aliquots at −80°C until further analysis. At the day of analysis, aliquots were thawed. Measurement of serum clinical biochemistry analytes was performed on the Olympus AU400 analyser except cortisol and TNF-alpha. The protocols for the use of Olympus System Reagents (OSRs) and other commercial reagents were in reference of previous studies (16, 17). Quality control protocols were based on the daily quantification of two control sera of low and high concentration (Control serum I and Control serum II, Beckman Coulter). For cortisol and TNF-alpha in piglets, two commercial ELISA kits were used: Cortisol ELISA Kit (DRG, Marburg, Germany) and Porcine TNF-alpha Quantikine ELISA Kit (R&D Systems, Abingdon, UK). The analytes, methods, reagents, and coefficients of variation (CV%) are shown in Table S3.

The RI establishment was in accordance with American Society for Veterinary Clinical Pathology (ASVCP) guidelines (18). These guidelines mirror the CLSI recommendations. All RIs were generated by Reference Value Advisor (19), an add-in program in Excel (Office 365, Microsoft, Redmond, WA, USA). For population amounts over 120, non-parametric RIs were directly used since this method doesn't assume any specific shape for data distribution. For populations <120 (24 h of age for calves and piglet samples), non-parametric RIs were not able to be computed. The criteria of outlier removal are following: (a) For interquartile range (IQR) = Q₃(third quartiles)—Q₁(first quartiles), values that exceed interquartile fence set at Q₁−3^*IQR and Q₃+3^*IQR are considered outliers; (b) Without removing certain outliers, if Reference Value Advisor computes result with good distribution and/or symmetricity, then the outliers are retained; (c) If no normal distribution and/or symmetricity are observed even outliers are removed and data are power transformed, such population is calculated with Cook's distance and Cook's outliers are detected and deleted. After the removal of outliers, populations were recalculated for RIs. Once the RIs were established, one-way analysis of variance (ANOVA) and Tukey's HSD with Bonferroni correction test, boxplots with difference annotations and distribution histograms were made by “R” (20) under RStudio environment (21) using the combinations of “multcompView” (22), “nortest” (23), “ggplot2” (24), “gridExtra” (25), and “ggpubr” (26) packages and values that exceed interquartile fence set at Q₁−3^*IQR and Q₃+3^*IQR were considered outliers.

It was noticed that the all enzymes (ALT, AST, GGT, GPx, and SOD) not only showed the highest concentrations at 24 h of age, but also significantly higher than any other times of age. The high concentration in newborn animals is an indication that these enzymes are absorbed from colostrum, at least in the case of GGT and AST (9, 32). The most extreme elevation occurred with GGT, which was 20 folds higher at 24 h than at 2 w and further decreases at 5 w and 7 w. Even though the GGT level is relatively low in adult cow's serum, its level in colostrum can peak-up to ~800-fold greater than in the serum of the same cow (33). GGT can increase its activity in calves' serum as early as 6 h after colostrum intake (intake at birth) (34) and similar to our results, it has been described that it takes approximately up to 5 weeks for GGT to decline to low level (adult's level) after the intake of colostrum (3, 11, 33). Other studies found that lower concentrations of GGT in new-born calves are linked with much lower levels of globulins/total protein and higher risk of death (33, 35); moreover, one study showed 0% sensitivity of GGT when it comes to the detection of hepatic diseases before 6 w old (36) since the highly concentrated colostrum-derived GGT presented in the serum can mask the level of GGT that is derived from liver damage. Indeed, a better characterization of GGT levels for new-born calves may help to monitor calves' health to prevent early-age death but not used to detect hepatic diseases. It is also relevant to notice the high individual variability of GGT values in 24 h old calves, probably indicating differences in colostrum intake, absorption and/or metabolization.

After 24 h of age, ALT's level reached its lowest at 2 w and gradually went uphill afterwards while AST stayed relatively steady and low levels from 2 to 7 w. However, at 7 w of age, both levels of ALT and AST were still lower compared with adult's RI. Other studies with longer terms showed that AST reaches adult's level around 8–10 w of life (9, 10). According to previous study along with GGT and other enzymes (36), AST's sensitivity to liver damage is 80% for calves <6 w old thus AST alone should not be considered clinical useful in diagnosing hepatic diseases. In addition, despite the statistical differences in AST throughout 2–7 w, none of their pair-wise comparison fold changes were larger than 1.15 or <0.85, thus biologically these were not considered major differences. For ALT, its characterization in cow is seldom studied because is not so commonly used for hepatic disease diagnosis since its activity is low and isn't massively liberated into serum during hepatic diseases in large animal (2).

SOD and GPx also showed rather stable levels from 2 to 7 w. Unfortunately, we didn't find any RIs of GPx for adult; and for SOD only RI of cow during transition period (60 d before until 100 d after milk) was found. Both mineral dependent enzymes [Mn, Cu and Zn for SOD (37) and Se for (2, 38)] are important enzymes whereas SOD can dismutate two O2- molecules to H₂O₂ and O₂ while GPx can catalyze H₂O₂ to H₂O (2). Study of SOD in RBC suggested that there is no such variation in the activity between cattle and human (39). Even though there are very few information about SOD and GPx, a better characterization may contribute to the monitor of minerals and oxidation in animal's body.

NEFAs and TGs were also elevated at 24 h because colostrum is rich in fat (40); their concentrations dropped at 2 w compared to 24 h and started increasing moderately with age until 7 w. While the level of NEFAs is almost comparable with adult's RI roughly from 5 w, TG's level is still beyond the adult's RI upper limit at 7 w. TGs can richly appear in chylomicrons, which are secreted by the intestine after fat-containing meal in preruminant calves. The TGs from chylomicrons can be hydrolyzed to NEFAs and then be transported to tissue for fat storage or for oxidation to produce energy (41); on the other hand, serum NEFAs can increase in case of fasting or negative energy balance. Indeed, the growth of calf requires a large amount of energy from fat and evidences have already shown that extra fat can increase BW gains (42, 43). An adult's-like RI in NEFAs from 5 w and a rather higher levels of TGs throughout the study indicate that the calves demanded high amount of energy for growth and no negative energy balance was noticed. Another form of TGs is very-low-density lipoprotein (VLDL) but it is generally low in bovine compared with in human (41).

Cholesterol had a very similar trend with one study (9) where the lowest concentration was found at 24 h and the concentrations started to increase smoothly. Even though it was previously described that colostrum contains higher cholesterol concentration compared with milk (44) and the serum cholesterol concentration is proportional to the colostral cholesterol concentration (45), the increase in total cholesterol along with increases in NEFAs and TGs probably indicate a major lipid mobilization within tissues after 24 h.

Glucose showed significant drop throughout age. The glucose supply from mother in neonate stops abruptly immediately after birth and the suckling neonate is entirely dependent on glucose supply via colostrum/milk, meanwhile the neonate needs to rapidly turn on pathways of endogenous glucose production as well as for fatty acid oxidation to maintain its organs and cells' functionality (46). The decrease of glucose and increase of NEFAs and TGs during the first 7 w may be the consequence of the high fat and relatively low carbohydrate content of colostrum/milk (47, 48). Despite the constant fall of serum glucose throughout the study, the RIs obtained from last measurement was still higher than the adult's one.

Moreover, creatinine, TP and urea also decreased throughout age until 7 w. Creatinine is a breakdown product of creatine (Cr) and creatine phosphate (PCr) in muscle and is cleared by the kidney. In human neonatology, it was well explained that the kidneys in fetus don't play a major role in maintaining fetal homeostasis (49, 50); also, the high blood concentration of creatinine comes from maternal transfer and endogenous Cr and PCr degradation and it does not indicate fetal renal failure. It was noticed that at 2 and 5 w the median creatinine levels were around the bottom line of the adult's RI while at 7 w it was below [median and SD at 7 w was 0.66 ± 0.12 mg/dL and adult's RIs are 1.0–2.0 mg/dL according to Kaneko et al. (2) and 0.9–1.4 mg/dL according to Pérez-Santos et al. (11)], comparing to another similar study (11), where mean creatinine laid within normal adult's RI from 6 d until 90 d of age. Yet we didn't find any other issues related to the muscle or health in these calves and a possible reason of this is due to diet/season/genetic variabilities between the farms. TP was also elevated at 24 h due to passive transfer of protein which mainly contains immunoglobulins from colostrum (32, 51, 52). This result is similar to those were reported in previous studies (11, 53). Urea had a similar change pattern to the TP along the time. It is the product of protein breakdown (52) and is synthesized by the liver and finally excreted by the kidney, which is directly influenced by the total protein level.

A special attention was paid to markers of stress since assessing a good welfare status is one of the main problems in calf production. The levels of haptoglobin didn't change along the time, similar to one previous study (54) and are lying within the health adult's RI range (10 mg/dL) (27). It is important to have specific RIs for haptoglobin in young calves since it is an acute-phase protein used as biomarker of inflammatory and infectious diseases (55), very common in young calves. The individuals with high haptoglobin values identified as outliers may be individuals with subclinical diseases, which were recognized as healthy in a clinical inspection. The antioxidant enzymes GPx and SOD, markers of oxidative stress (56) had higher values at 24 h, but the RIs are similar afterwards.

As it was previously mentioned, on week 2 we found trial-biased distributions (Figure S1) on AST and cholesterol, where T3 showed lower concentration range compared with others. This reason may be because the average age in T3 is ~2 d older than other two. This difference indicates that only 2-days age difference can cause mean concentration difference in some parameters, thus special care should be taken when comparing local clinical values with RIs in literature.

No major changes were noticed on hepatic enzymes (ALT, AST, and GGT), glucose and TP. Cholesterol, NEFAs and TGs showed their highest values just after weaning (0 d). This may be due to the sudden dietary change from high fat content of milk to less-digestible, more-complex solid feed that piglets eat after weaning (57).

Throughout the ages, all medians of urea were within adult's RI while all medians of creatinine were <0.8 folds to the adult's RI lower fence. Both analytes showed highest concentrations at 7 d post-weaning. The increase in urea and creatinine only at 7 d post-weaning may be due to transient dehydration after the dietary change. These pattern changes are very similar to one study between 5-day-old unweaned piglets and 30-day-old post-weaning piglets (weaned at 28 d of life) supplemented with iron in solid diet (12). However, since this report only measured 5 d and 30 d, different from our opinion, the author's explanation of the increases was due to the increase of muscle mass and helped by the gradual supplementation of enriched solid diet.

Special interest has been devoted to markers of stress and inflammation. Thus, cortisol was measured as the paradigmatic stress hormone and found higher just after weaning, similar to other studies (58–60), this probably is due to the stress associated to separation of the mother and change of diet; it's also interesting to notice that from 7 d post-weaning, the mean concentrations of piglets are comparable to healthy adults' one (31), which suggests that in this study the diet change only caused a transient stress to the animals. The acute phase proteins haptoglobin and Pig-MAP were also measured; whereas haptoglobin RIs were not influenced by age and weaning, Pig-MAP was higher at 3 weeks-old just after weaning. Absolute values for haptoglobin were similar to those reported at 3-week-old (13). In that study, reference values for both acute phase proteins were established for pigs older than 4 weeks. Pig-MAP is also considered a stress marker, which increases in stressful conditions like transport (61–63). Finally, the cytokine TNF-alpha is frequently used as a marker of infection, especially in gastrointestinal disorders which often affect piglets at weaning (31, 64, 65). For this reason, it is interesting to have RIs available in healthy animals. A moderate increase in TNF-alpha was shown at 7 and 14 days after weaning, probably due to the change of diet. This increase in serum may be secondary to the transient response in gene expression of inflammatory cytokines in the gut after weaning (66).

Similar to the calves, creatinine from 0 d post-weaning (3 w of age) was also comparable to the adult's RIs. GGT's RI is also comparable to adult's one from 7 d of age while for calves it takes 5 weeks to drop into adult's RI (2).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.