Effects of a mixture of phytobiotic-rich herbal extracts on growth performance, blood metabolites, rumen fermentation, and bacterial population of dairy calves

Forty-eight newborn Holstein dairy calves [40 ± 3.4 (SD) of kg of body weight (BW); 24 females and 24 males] were used in a completely randomized design to investigate the effects of a mixture of phytobiotic-rich herbal extract (Immunofin, IMPE) incorporated into milk on performance, ruminal fermentation, bacterial population, and serum biochemical metabolites during the preweaning period. Calves had free access to calf starter and clean water from d 6 until weaning. The treatments were the control (CON; without additive) and IMPE at 4, 8, and 12 mL/d. T`he treatments had no significant effect on total dry matter intake, weight gain, and BW at weaning. The incidence of diarrhea was lower in calves fed 8 mL of IMPE/d compared with CON. At weaning, body measurements (except for front leg circumference) were not affected by IMPE treat-ment. Relative to the CON group, front leg circumference was significantly decreased by IMPE supplementation. Serum IgG concentration was not significantly increased by IMPE supplementation compared with the CON group. Triglyceride concentration decreased in calves receiving 4, 8, and 12 mL/d of IMPE compared with the CON groups. In contrast to the CON group, serum albumin and total serum protein concentrations increased with IMPE supplementation. Calves receiving 4 mL/d of IMPE had a greater abundance of total bacteria, Ruminococcus albus , Ruminococcus flavefaciens , and Fibrobacter succinogenes compared with the other treatments. Molar proportions of acetate increased in calves fed IMPE (at 12 mL/d) compared with calves fed CON. Ruminal N-NH 3 concentrations decreased linearly with the increase in IMPE supplementation. The results of the present study suggest that the addition of IMPE to milk may improve some health and immunity conditions, blood metabolite concentrations, and increase the abundance of some cellulolytic bacteria in the rumen of Holstein dairy calves. The use of IMPE may be an alternative to feeding antibiotics at subtherapeutic concentrations to improve calf health and immunity status.


INTRODUCTION
Restrictions on the use of antibiotics in animal feed have stimulated several research studies to identify alternatives (Cross et al., 2007;Miguel, 2010;Wu et al., 2020).Studies have been conducted on various feed additives as alternatives to antibiotics, including probiotics, immunostimulants, antimicrobials, antioxidants, and bioactive plant compounds.Plant phytochemical derivatives (HPD) contain a variety of antimicrobial, antifungal, and immunostimulant compounds (Miguel, 2010).
Some herbal extracts (such as Crina Ruminants, Fructus Ligustri Lucidi, Radix Astragali, and Radix Codonopsis) have been confirmed to improve the efficiency of rumen fermentation in cattle (Beauchemin and McGinn, 2006;Qiao et al., 2013).However, conflicting results have been shown regarding the effect of herbal extracts on dairy calf performance and health (Akbarian-Tefaghi et al., 2018;Liu et al., 2020;Reddy et al., 2020).The differences are most likely due to the extraction method, the amount of supplementation, the type of plant, differing diet composition, and the animal conditions.
Calves supplemented with mixed essential oils (eucalyptus oil, menthol crystal, mint oil) in milk replacer demonstrated an increase in growth performance and better general health, as well as reduced antibiotic usage before weaning (Soltan, 2009).However, Santos et al. (2015) reported that essential oils (a mixture of carvacrol, cineole, cinnamaldehyde, pepper oil) Effects of a mixture of phytobiotic-rich herbal extracts on growth performance, blood metabolites, rumen fermentation, and bacterial population of dairy calves in milk replacer (at 400 mg/kg of milk replacer) or supplemented in a combination of milk replacer (200 mg/kg) and starter feed (200 mg/kg) had no effect on growth performance, fecal scores, and gut microbiota of dairy calves.According to Seifzadeh et al. (2017), the addition of herbal plant mixtures increased ADG, starter feed intake, and serum BHB concentration in dairy calves.Thus, in addition to the supplementation method (in liquid or solid feed), the type of plant from which the extract is obtained also affects the results.In addition, heating during calf starter production can lead to oxidation, evaporation, or conversion of essential oils (EO) secondary compounds (Turek and Stintzing, 2013).Therefore, EO supplementation with liquid feed can maximize quality conservation and its expected effects.The esophageal groove closure reflex in young calves prevents milk from entering the rumen (Kaba et al., 2018).Failure of the esophageal groove can be caused by pathological conditions, drinking cold milk, feeding milk with buckets, and stressors (Lateur-Rowet and Breukink, 1983;Gentile, 2004;Kaba et al., 2018).Also, milk containing antibiotics has been found to alter the rumen microbiota (Li et al., 2017), which is particularly the case when feeding bucket milk, as the gastrointestinal ecosystem is open and integrated (Savage, 1977).With this in mind, we speculated that absorbed plant secondary compounds might pass into the rumen lumen after absorption from the blood (Shen et al., 2018), and thus the inclusion of EO in the liquid feed might alter microbial diversity and activity in the rumen.
Citrus EO, peppermint, yarrow, and coneflower extract have attracted considerable attention in several studies among herbal bioactive compounds for their antimicrobial, antiinflammatory, and antioxidant activities against pathogenic bacteria (Rehman et al., 1999;Hsouna et al., 2017;Mohammadhosseini et al., 2017).A commercial blend of phytobiotic-rich herbal extracts (known as IMPE) and the immunostimulant properties attributed to it are expected to improve the health and growth performance of preweaning calves.We hypothesized that IMPE supplementation could lead to improved health, immunity, and growth performance of dairy calves by increasing dietary intake of bioactive phytochemicals.However, the effects of HPD on performance and health are variable, depending on the dosage and the main secondary compounds (Busquet et al., 2006).Therefore, the study objective was to investigate the effects of different doses of IMPE (at 4, 8, and 12 mL/d) supplemented in whole milk on growth performance, ruminal fermentation, and bacterial population as well as biochemical and immunological parameters in the serum of dairy calves.

Animal and Experimental Design
This study was conducted in Taherabadi dairy farm (34°31′ N 48°01′ E), Kermanshah, Iran, during the period from May 12 to August 4, 2017.The experimental protocols were approved by the Animal Care and Use Committee of the University of Kurdistan (IACUC # 2017001) according to the Iranian Council of Animal Care (1995).Forty-eight newborn Holstein calves (40 ± 3.4 kg of BW; 24 females and 24 males) were used in a completely randomized design.After calving, calves were immediately separated from their dams and dried, then weighed and housed in individual pens (2.5 × 1.2 × 1.6 m) bedded with dry wheat straw until 61 d of age.Calves were fed 3 L of colostrum (with average Brix value 24%) within the first 6 h of birth, followed by 2 L for a total of 5 L in the first 12 h of life.Calves received 4 L of transition milk for d 3 of life in open buckets, twice daily (8 h and 20 h).Thereafter, calves received 4 L/d from d 3 to 10, 5 L/d from d 11 to 56, 2.5 L/d from d 57 to 60, and then calves were weaned.A sample of whole milk offered was taken weekly and analyzed for protein (3.15 ± 0.07%), fat (3.5 ± 0.8%), lactose (4.7 ± 0.2%), and solids-not-fat (9.1 + 0.8%) and bacterial count using automatic infrared analysis (Milkana Kam 98-2A, Bulteh).The total bacterial count of milk during this study ranged from 98,000 to 125,000 cfu/mL.Calves had free access to calf starter and clean water from d 6 until weaning.Table 1 shows the ingredients and chemical composition of calf starter.Calves were randomly assigned to experimental treatments (12 calves per treatment).The treatments were as follows: control (CON; without additives) and a commercial herbal extract mixture (IMPE; in 4, 8, and 12 mL/d).The IMPE treatments were added to the daily milk in equal amounts at each meal during the experimental period (from d 3).The IMPE (Immunofin, Pars Imen Daru Co) contained coneflower (Echinacea angustifolia), yarrow (Achillea millefolium), peppermint (Mentha piperita), and citrus EO.Table 2 shows the results of GC/MS (gas chromatograph: Agilent, 7890B; mass spectrometer: Agilent, 5977A; column: B 35 ms, 30 m × 0.25 mm, 0.25 µm thickness) of the analysis of IMPE.The IMPE was a branded product proposed for modulation of the immune system, health, and performance in dairy calves.

Sampling
To calculate ADG, BW was measured using a digital scale on d 15, 30, 45, and 60 of the experimental Jahani-Azizabadi et al.: PHYTOBIOTIC EXTRACT AND CALF PERFORMANCE periods.The average individual starter intake was measured in the 15-d intervals (except for first recording time from d 6-15).Starter feed samples were collected twice from the beginning to the end of the experimental period and then stored frozen at −20°C.Weekly milk samples were collected twice daily from the beginning to the end of the experimental period.Then, the daily samples were mixed and analyzed to determine the milk composition.At weaning, body measurements including withers height, body length, heart girth, front leg circumference, and leg circumference were measured using a flexible plastic meter.Body length was the length of the animal from the point of the shoulders to the pin bone.Heart girth measured as the chest's circumference, and withers height was the distance from the base of the front feet to the withers.The fecal score [from 1 (normal) to 4 (several problems)] was determined daily based on the physical shape of the feces.To determine diarrhea, occurrence scores were categorized as number of calves with a fecal score <3 [firm (score = 1, normal), soft (score = 2, mild diarrhea but not treated)] that were considered healthy and calves with a fecal score ≥3 [liquid (score = 3, moderate diarrhea), watery (score = 4, severe diarrhea)] that were considered diarrheic and required treatment (Kargar et al., 2018).Symptoms of severe diarrhea include calves that lie down and wake only when called, a returning skin tent >6 s, severely sunken eyes, white and dry gums, and 8 to 10% dehydration.Moderate diarrhea is characterized by reduced sucking reflex, the return of the skin tent within 2 to 6 s, sunken, slightly depressed eyes, and 6 to 8% dehydration.Calves with moderate diarrhea were treated with oral rehydration solution (3 L/calf per day).All calves that showed severe diarrhea were treated with neomycin (neomycin sulfate 500 mg; 2 bolus/calve per day), enrofloxacin (enrofloxacin 10%; 1.5 mL/calf per day), and flunixin meglumine (flunixin 5%; 4 mL/calf per day) for 3 consecutive days.In addition, calves with severe diarrhea were treated with oral rehydration solution after receiving injectable hypertonic solution (sodium chloride 7.2%; 500 mL/calf per day) to make sure the calves drank enough of the rehydration solution.
Blood samples were collected from the jugular vein 4 h after morning milk feeding on the last 2 consecutive days before weaning (d 58 and 59) using vacutainer-clot activator tubes.Serum was separated after 4 h of residence at 4°C and stored in a microtube at −20°C until analysis.On d 60, rumen fluid samples were collected 4 h after morning feeding using a stomach tube and a vacuum pump (JB industries DV-3E Eliminator), and the samples were centrifuged at 3,000 × g for 15 min at 4°C.Then, 1.5 mL of the supernatant was collected and mixed with 375 µL of 25% orthophosphoric acid and stored in a microtube at −20°C until analysis to determine the VFA concentration in the rumen.In addition, 5 mL of centrifuged rumen fluid was collected and acidified with 5 mL of 0.2 N HCl to determine the N-NH 3 concentration in the rumen fluid.Rumen contents were collected and stored at −80°C until DNA extraction and analysis for the relative abundance of ruminal bacteria by quantitative real-time PCR (qPCR).

Chemical Analyses
The starter samples were mixed and oven-dried at 60°C for 72 h.A sample was then taken and ground so that it passed through a 1.5-mm sieve.The feed samples were analyzed for DM, CP, and ash according to standard procedures (AOAC International, 2000).Ash-free NDF and ADF content of the diets were determined according to the method recommended by Van Soest et al. (1991) andAOAC International (2000), respectively.Nitrogen concentration of feed samples and N-NH 3 concentrations in the rumen fluid were determined by the Kjeldahl method (Kjeltec 2300, Foss Tecator AB).Volatile fatty acid concentrations were determined using the gas chromatography method recommended by Jouany (1982;PU 4410;Philips Unicam) triglyceride, total serum protein (TSP), albumin, aspartate aminotransferase, and alanine aminotransferase were determined by enzymatic methods (Pars Azmun co).The concentration of BHB was determined with commercial clinical kits (Randox Laboratories) using a spectrophotometry system (JASCO, V-570) according to the manufacturer's instructions.Serum concentration of IgG was determined with special kits from ELISA (Pishtazteb co).

DNA Extraction and qPCR
The samples of rumen contents were thawed and mixed in 1.5-mL tubes containing 0.5-mm diameter glass beads.Samples were incubated on ice for 2 min and then vortexed to dislodge the microbes from the feed particles.Samples were then centrifuged at 2,000 × g for 10 min at 4°C, and total DNA was extracted from approximately 200 µL of the supernatant using a genomic DNA extraction kit (AccuPrep, Bioneer Corporation).Amplification and detection were performed in triplicate using an ABI Real-Time PCR system (Applied Biosystems 7300).The reaction was performed in a final volume of 25 µL of reaction mixture consisting of 12.5 µL of SYBR Green PCR Master Mix (Maxima SYBR Green/ROX qPCR Master Mix, K0221, Fermentas), 1 µL of DNA template, 0.5 µL of each primer (containing 10 pmol primers), and 11 µL of deionized water.Table 3 shows the species-specific PCR primers used in this study.The DNA amplification program was performed as described by Valizadeh et al. (2010).The relative abundances of bacteria were determined using the 2 −ΔΔCt method (Livak and Schmittgen, 2001).The fold change in DNA of specific bacterial species in calves fed the experimental treatments compared with the CON diet was calculated by normalizing the DNA of specific bacterial species to the total bacterial DNA in the experimental groups and relating this ratio to that of the CON group.Changes in specific bacterial species are expressed as a fold change in genomic DNA per 1 µL of extracted DNA compared with CON.

Statistical Analyses
For the response variables, including feed intake, BW, and ADG, a power analysis was performed to estimate sample size (Morris, 1999;Hintze, 2008) based on previously published values (DeVries and von Keyserlingk, 2009;Zhang et al., 2010;Miller-Cushon and DeVries, 2011).A power test analysis with α = 0.05 and power = 0.80 can detect biologically relevant treatment differences for growth performance to determine the target sample size.All data were tested for normal distribution using the UNIVARIATE procedure of SAS (the Shapiro-Wilk test) before analysis.Feed intake, BW, ADG, body measurements, and blood metabolites were normally distributed for all variables, but the remaining values were not normally distributed.The parameter that was not normally distributed was transformed using log 10 transformation.Data were subjected to ANOVA using the MIXED procedure of SAS (PROC MIXED, SAS 2003, SAS Institute Inc.) with time as repeated measures for starter feed intake, total DMI, and ADG.For the intake of starter feed and total DMI, we considered daily intake, BW, and ADG every 15 d as repeated measures.The model for starter feed intake, total DMI, BW, and ADG consisted of calf sex, treatment, time (sampling date), and treatment × time interaction as the fixed effects, and calf as the random effect.Body measurements, blood metabolite data, ruminal fermentation characteristics, and ruminal relative bacterial population were analyzed using the above model without the effect of time.The CONTRAST statement of SAS was used to test the linear, quadratic, and orthogonal contrasts (CON × IMPE) of IMPE inclusion in the milk (0, 4, 8, and 12 mL/d).Orthogonal coefficients for unequally spaced treatments were acquired using PROC IML of SAS.The GENMOD procedure was used to analyze diarrhea occurrence.The Tukey-Kramer adjustment was applied to account for multiple comparisons.The threshold of significance was set at P ≤ 0.05; trends were declared at 0.05 < P ≤ 0.10.

IMPE Compounds
Table 2 shows the main secondary compounds of the IMPE.In this experiment, 81.28% monoterpenes and 4.04% sesquiterpene compounds were identified in the IMPE.d-Limonene, δ-3-carene, and γ-terpinene were found to be the highest (55.56, 10.55, and 4.57%, respectively) of the secondary compounds of IMPE among the terpenoids.

Growth Performance
Weaning BW, ADG, total DMI, starter feed intake, and skeletal growth (except for front leg circumference) of dairy calves (Table 4) were not significantly affected by increasing IMPE supplementation.In contrast, previous studies reported positive effects of supplementation of EO and plant extracts on the ADG of calves (Jeshari et al., 2016;Liu et al., 2020).Similar studies investigating supplementation of a mixture of commercial EO have obtained different results for DM and starter intake.Swedzinski et al. (2019) and Froehlich et al. (2017) found that there were no significant differences in total DMI and starter intake in calves fed a mixture of EO [a blend of oregano and thyme EO, prebiotics (arabinogalactan), vitamins, probiotics, and microbial catalyst (0.4 g/kg of milk replacer)].In contrast, EO appears to have a more positive effect on the starter and total DMI when supplemented with calf starter (Jeshari et al., 2016;Liu et al., 2020).Liu et al. (2020) found that calves supplemented with EO (arabinogalactans, carvacrol, thymol, cineole, caryophyllene, p-cymete, terpinene, and cobalt lactate; 44.1 mg/kg) showed an increase in total DM and starter intake.In the present study, there was no significant effect of IMPE supplementation on starter intake (Fig- ure 1) and ADG (Figure 2).Froehlich et al. (2017) and Liu et al. (2015) observed that feeding EO at high doses negatively affected DMI and ADG.Although we expect milk to pass from the rumen to the abomasum due to the esophageal groove reflex, there is evidence that milk may have entered the rumen (Li et al., 2017).Therefore, the supplementation of antimicrobial compounds from milk may affect the rumen microbiota, as observed in the present study.In addition, the effects of EO on DMI could be due to the effects of its volatile and aromatic compounds on general feeding appetite or the effects on palatability of the feed when supplemented in the calf starter (Calsamiglia et al., 2007;Franz et al., 2010).The beneficial effects and biological effects of HPD in calves were associated with the doses, the main bioactive compounds, the state of use of HPD (the whole herb, the extract and the microencapsulated or the pure EO), the mode of administration (through the solid or liquid feed), and the time required to acquire the effects of HPD.

Health and Body Measurements
One of the limitations of the study was that rectal temperature was not recorded daily.Figure 3 shows the effect of treatments on the occurrence of diarrhea (score ≥3) during the preweaning period.In this study, the relative occurrence of diarrhea was lower in calves receiving 8 mL/d of IMPE compared with the CON group (Figure 3; P < 0.05).Compared with the CON group, IMPE supplementation resulted in a lower occurrence of diarrhea (50, 20, and 50% for the 4, 8, and 12 mL of IMPE/d per head versus 60% for the CON groups, respectively).In this study we observed that 3  calves from the CON group showed severe diarrhea and 3 showed signs of moderate diarrhea.Moreover, out of all calves (n = 36) receiving IMPE, only 3 animals had severe diarrhea, and 9 showed moderate diarrhea.The results of this study showed that IMPE (at 8 mL/d) has a high potential to decrease diarrhea occurrence in dairy calves.Similar studies have reported the antimicrobial effect of HPD in reducing diarrhea in newborn calves (Froehlich et al., 2017;Salazar et al., 2019).Salazar et al. (2019) noted a decrease in diarrhea frequency with the addition of 300 mg of Crina (a commercial mixture of thymol, guaiacol, vanillin, and limonene) per kilogram of calf starter.In contrast to our observation, Soltan (2009) reported a decrease in diarrhea and improvement in overall health when EO (Activo, contains microencapsulated carvacrol, cine oil, cinnamaldehyde, pepper oil resin, and mannan oligosaccharide) was supplemented through the calf starter instead of the milk replacer (Santos et al., 2015).
The main components present in IMPE are monocyclic (d-limonene and γ-terpinene) or bicyclic monoterpenes (δ-3-carene).Antibacterial activity of these bioactive components against Escherichia coli, Mycoplasma pneumoniae, Klebsiella pneumonia, and Clostridium sp. are confirmed in previous studies (Furneri et al., 2012;Perumalsamy et al., 2013), which decrease the frequency of diarrhea and respiratory infections in neonatal dairy calves.Monoterpenes are recognized as being growth inhibitors for gram-positive and gramnegative bacteria (Cristani et al., 2007).The monoterpene mode of action against pathogenic bacteria can be described by the permeability of the cell wall, damaging the cytoplasmic membrane, and interacting with intracellular contents (Nazzaro et al., 2013).Cristani et al. (2007) reported that monoterpene antimicrobial activity depends on their ability to affect penetrating the cell and interacting with intracellular target sites.
The main components contained in IMPE are monocyclic (d-limonene and γ-terpinene) or bicyclic monoterpenes (δ-3-carene).The antibacterial activity of these bioactive components against Escherichia coli, Mycoplasma pneumoniae, Klebsiella pneumonia, and Clostridium sp. was confirmed in previous studies (Furneri et al., 2012;Perumalsamy et al., 2013), and they reduced the incidence of diarrhea and respiratory infections in newborn dairy calves.Monoterpenes are known to be growth inhibitors of gram-positive and gram-negative bacteria (Cristani et al., 2007).The mode of action of monoterpenes against pathogenic bacteria can be described by cell wall permeability, cytoplasmic membrane damage, and interaction with intracellular contents (Nazzaro et al., 2013).Cristani et al. (2007) reported that the antimicrobial activity of monoterpenes depends on their ability to penetrate the cell and interact with intracellular target sites.Further studies with more animals considering duration of diarrhea treatment are needed to confirm these results.
There were no significant differences in the measurements of withers height, body length, and leg circumference of calves at weaning (Table 4).However, calves receiving 12 mL/d of IMPE had a smaller (P < 0.01) front leg circumference relative to the others treatment.In contrast to the CON group, the front leg circumference was significantly decreased (P < 0.01) by the IMPE supplementation.The results showed that an increase in IMPE supplementation was associated  with a decrease in heart girth and leg circumference (Table 5).There are various observations on the effect of supplementation of herbal extracts on body measurements of newborn ruminants.In agreement with this observation, Santos et al. (2015) showed no changes in body measurements with supplementation of EO.
In contrast to our results, Froehlich et al. (2017) and Liu et al. (2020) observed that body measurements were higher in calves fed 0.5 g/d of a mixture of EO.There is a positive correlation between body measurements and the growth performance of calves.This inconsistency between results could be caused by the amount of EO supplementation (4-12 mL/d vs. 0.5 g/d), EO source, and method of use (inclusion in milk, starter, or both).

Blood Biochemical Metabolites
Blood concentrations of IgG, BHB, aspartate aminotransferase, alanine aminotransferase, globulin, glucose, cholesterol, and urea nitrogen of dairy calves were not affected by IMPE supplementation (Table 5).The results of this study showed that serum triglyceride concentration decreased linearly in dairy calves receiving IMPE (P < 0.05).Comparison of treatments explained that triglyceride concentration significantly decreased in calves receiving 4, 8, and 12 mL/d of IMPE compared with CON (P < 0.05).However, serum IgG concentrations did not increase significantly by the different amounts of IMPE (4, 8, and 12 mL/d) relative to the CON group (13.28, 12.13, 12.38 vs. 10.70 mg/mL, respectively).Serum albumin and TSP concentrations also increased linearly in dairy calves with increasing IMPE supplementation (P < 0.05).In addition, calves fed 12 mL/d of IMPE had higher albumin concentrations at weaning than CON (P < 0.05).Although serum albumin and TSP concentrations were lower in the CON groups (P < 0.05), Froehlich et al. (2017) reported that the inclusion of a mixture of EO (containing thymol, caryophyllene, p-cymene, cineole, terpinene, and carvacrol) in the calf diet resulted in a nonsignificant increase in serum IgG concentration, similar to our results.
The amount of serum immunoglobulin may reflect the regulatory process of humoral immunity against pathogenic organisms.B cells are responsible for the production of specific globulins such as IgG (Gelsinger and Heinrichs, 2017).Previous studies confirmed that IgG concentration was lower in calves suffering from severe diarrhea than in healthy calves (Villarroel et al., 2013).Wang et al. (2015) suggested that serum IgG concentration and diarrhea intensity were negatively correlated in dairy calves.
The linear decrease in serum triglyceride concentration of dairy calves receiving IMPE contradicted the observation of Seifzadeh et al. (2017) that serum triglyceride concentration was not significantly affected by the inclusion of a mixed medicinal plant in the starter (1.5 and 3% of the starter DM).The greater serum triglyceride concentrations in the CON group than other groups may be associated with diarrhea occurrence.It has been reported that triglyceride concentration increased in dairy calves with diarrhea due to cytokine activation in diarrheic calves (Khovidhunkit et al., 2000).In this study, calves fed IMPE had less diarrhea, which was associated with reduced serum triglyceride concentration.

Ruminal Fermentation and Microbial Population
When phytochemical derivatives of herbal extracts are included in the calf's diet, it is important to study the effects on the microbial population and diversity in the rumen because of their antimicrobial properties.Table 6 shows the effects of the experimental treatments on the relative abundance of rumen bacteria of calves at weaning.Calves receiving 4 mL/d of IMPE had a greater abundance of total bacteria, R. albus, R. flavefaciens, and F. succinogenes compared with CON (P < 0.05).Independent contrast showed no significant differences between the CON and IMPE groups in the relative abundance of F. succinogenes.The relative abundance of total and investigated cellulolytic bacteria decreased linearly (P < 0.01) with increasing level of IMPE supplementation.
It seems that herbal extracts from different sources may have stimulatory or inhibitory effects on fibrolytic bacteria due to their nonvolatile compounds (Khovidhunkit et al., 2000), and pure EO may have inhibitory effects on gram-positive bacteria due to main bioactive compounds (Cristani et al., 2007;Khorrami et al., 2015).We expect milk passes from the rumen due to the esophageal groove reflex, and there is evidence that milk containing antibiotics affects the rumen microbiota (Li et al., 2017).This is expected because the gastrointestinal ecosystem is open and integrated (Savage, 1977).In addition, antibiotics of plant secondary compounds may be transferred from blood into the gastric lumen after absorption.The results of Li et al. (2017) on the effects of antibiotics in milk on microbial diversity in the rumen of newborn calves confirm our observation.Santos et al. (2015) reported that the cellulolytic bacteria population was not affected by supplementation of a mixture of EO (400 mg/d containing oligosaccharides and cinnamaldehyde, carvacrol, cineole, and pepper oil) in milk replacer or starter.The inconsistency of the results could be related to the dosage (4-12 mL/d vs. 0.5 g/d).Fibrobacter succinogenes, R. albus, and R. flavefaciens were considered the major cellulolytic rumen bacteria in this study.Gram-negative bacteria are thought to be more resistant to EO than gram-positive bacteria due to a protective outer membrane surrounding the cell wall (Burt, 2004).Fibrobacter succinogenes is a gram-negative and R. flavefaciens and R. albus are gram-positive bacteria among these 3 cellulolytic bacteria.Therefore, the results of this study showed that IMPE is a potent rumen manipulator that affects gram-negative and gram-positive rumen microbes.The relative abundance of cellulolytic bacteria decreased (P < 0.05) when 8 and 12 mL/d of IMPE were fed compared with 4 mL/d.This could be due to an increase in rumen input of some secondary compounds, which have high antimicrobial activity, due to an increase in IMPE consumption.
Compared with CON, total VFA and molar proportions of propionate, butyrate, and isovalerate, as well as the acetate-propionate ratio in rumen fluid, were not affected by the treatments (Table 7).The molar proportions of acetate increased linearly with the increase in IMPE supplementation (P < 0.05).In contrast to CON, the molar proportion of acetate increased significantly (P < 0.05) in calves fed 12 mL of IMPE/d.Our observation showed the molar proportion of propionate was not affected by IMPE supplementation compared with CON.Molar proportion of propionate quadratically decreased (0.05 < P ≤ 0.10) with an increase in IMPE supplementation.However, IMPE supplementation at 12 mL/d decreased (P < 0.05) the molar proportion of propionate and increased the molar proportion of acetate.Ammonia nitrogen (N-NH 3 ) concentrations decreased linearly (P < 0.01) with increasing IMPE supplementation.Calves fed IMPE had lower rumen N-NH 3 concentrations compared with the CON group.
The results of this study for the molar proportions of VFA explained that IMPE supplementation increased the abundance of gram-positive bacteria in the rumen.However, IMPE at 8 and 12 mL/d had no significant effect on gram-positive bacteria in the rumen, relative to CON.In general, gram-positive rumen bacteria are acetate and butyrate producers, whereas gram-negative bacteria are usually propionate producers (Stewart, 1991).In contrast to the results of the present study, Akbarian-Tefaghi et al. (2018) reported that ruminal VFA and N-NH 3 concentrations were not affected by supplementation of a commercial mixture of EO (containing limonene, thymol, and carvacrol) in dairy calves after weaning.Santos et al. (2015) reported a similar observation.Poudel et al. (2019) also observed an increase in propionate concentration in dairy calves fed a diet supplemented with a mixture of EO (thymol, caryophyllene, p-cymene, cineole, terpinene, and carvacrol).Inconsistent results from different studies on the effects of EO on rumen fermentation depend on diet composition, environment, major bioactive compounds, and doses.It is possible to observe an improvement in nitrogen metabolism by decreasing ruminal N-NH 3 concentration and serum urea nitrogen concentration by IMPE supplementation.Akbarian-Tefaghi et al. (2018) reported that N-NH 3 concentration was not affected by the addition of EO.Meanwhile, Santos, et al. (2015) observed that EO supplementation increased ruminal N-NH 3 concentration in dairy calves compared with our observations.McIntosh et al. (2003) and Jahani-Azizabadi et al. (2011) reported that EO compounds have an inhibitory effect on N-NH 3 production by hyperammonia-producing bacteria.The success of IMPE in affecting ruminal N-NH 3 concentration was consistent with the numerically decreased serum urea nitrogen concentration in calves fed different amounts of IMPE.The results of the present study suggested that IMPE is an effective natural feed additive for improving the protein metabolism of dairy calves.

CONCLUSIONS
Contrary to our hypothesis, IMPE supplementation did not affect growth performance, rumen fermentation, and serum IgG concentration of dairy calves.In addition, IMPE might serve as an alternative to antibiotics to reduce the risk of health disorders in young calves.This study was subjected to some limitations, including a relatively small sample size as well as the lack of recording of daily starter intake and rectal temperature.Further studies with more animals considering daily records are needed to confirm these results.

Figure 1 .
Figure 1.Average starter intake of calves fed milk without (CON) or with a blend of phytobiotic-rich herbal extract (IMPE) at 4, 8, and 12 mL/d in 15-d intervals.Data are LSM ± SEM.

Figure 2 .
Figure 2. Average daily weight gain of calves fed milk without (CON) or with a blend of phytobiotic-rich herbal extract (IMPE) at 4, 8, and 12 mL/d in 15-d intervals.Data are LSM ± SEM.

Figure 3 .
Figure 3. Diarrhea occurrence (proportion of sick calves) during experimental period in calves fed milk without (CON) or with a blend of phytobiotic-rich herbal extract (IMPE) at 4, 8, and 12 mL/d.Healthy: fecal score <3, Diarrhea: fecal score ≥3.Groups with different letters (a, b) are significantly different (P < 0.05).Data are LSM.The closed shaded region shows proportions of diarrhea calves and dotted region shows proportions of healthy calves.

Table 5 .
Effects of herbal extracts (Immunofin, IMPE) on blood metabolites of dairy calves (n = 12 per group)