Saccharomyces cerevisiae boulardii CNCM I-1079 affects health, growth, and fecal microbiota in milk-fed veal calves

J. Dairy Sci. 102:7011–7025 https://doi.org/10.3168/jds.2018-16149

© 2019, The Authors. Published by FASS Inc. and Elsevier Inc. on behalf of the American Dairy Science Association®. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Saccharomyces cerevisiae boulardii CNCM I-1079 affects health, growth, and fecal microbiota in milk-fed veal calves C. Villot,1 T. Ma,1,2 D. L. Renaud,3 M. H. Ghaffari,1* D. J. Gibson,1 A. Skidmore,4 E. Chevaux,4 L. L. Guan,1 and M. A. Steele1,5† 1

Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, T6G 2P5, Canada Feed Research Institute, Chinese Academy of Agricultural Sciences, Key Laboratory of Feed Biotechnology of the Ministry of Agriculture, Beijing, 100081, China 3 Department of Population Medicine, University of Guelph, Guelph, ON, N1G 2W1, Canada 4 Lallemand Animal Nutrition, F-31702 Blagnac, France, and Milwaukee, WI 53218 5 Department of Animal Biosciences, University of Guelph, Guelph, ON, N1G 2W1, Canada 2

ABSTRACT

The objective of this study was to investigate the effect of one specific strain of yeast, Saccharomyces cerevisiae boulardii CNCM I-1079 (SCB), on the growth performance, health, and fecal bacterial profile of veal calves. A total of 84 animals were enrolled in an experiment at a commercial veal farm for a total of 7 wk. Calves were fed twice a day with a milk replacer meal during the entire experiment and were randomly assigned to receive daily either SCB supplementation (10 × 109 cfu/d) or a placebo (CON). Individual feed intake and body weight were monitored on a daily and weekly basis, respectively. Fecal samples were collected at arrival to the veal facility (wk 0) and additional samples were taken on d 14 (wk 2) and d 49 (wk 7). These samples were subjected to 16S rRNA gene amplicon sequencing using Illumina MiSeq (Illumina Inc., San Diego, CA) to examine the bacterial profiles and real-time quantitative PCR to quantify Saccharomyces cerevisiae and specific bacterial groups. The significant increase of S. cerevisiae in the feces of SCB calves at wk 2 and 7 compared with wk 0 (respectively 1.7 × 107, 1.2 × 107, and 2.2 × 105 copy number of S. cerevisiae/g of feces) indicates a good survival of that yeast strain along the gastrointestinal tract. Supplementation of SCB did not improve overall growth performance with regard to average daily gain (ADG), final body weight, and feed intake. Nevertheless, a total of 69.1% of nonsupplemented calves had diarrhea and 28.6% experienced severe diarrhea, whereas 50.0% of the calves supplemented with SCB had diarrhea and 9.5%

Received December 11, 2018. Accepted March 29, 2019. *Current address: Institute of Animal Science, Physiology and Hygiene Unit, University of Bonn, 53115 Bonn, Germany. †Corresponding author: masteele@​uoguelph​.ca

experienced severe diarrhea. With respect to antibiotic use, 89.7% of the diarrheic calves recorded in the CON group were treated, whereas only 66.7% of the SCB diarrheic calves received an antibiotic. In addition, diarrheic calves supplemented with SCB maintained an ADG similar to nondiarrheic animals, whereas the CON diarrheic calves had a significantly lower ADG in comparison with nondiarrheic CON calves. Fecalibacterium was the most predominant bacterial genus in fecal samples of nondiarrheic and diarrheic calves supplemented with SCB, whereas fecal microbiota was predominated by Collinsella in diarrheic calves from the CON group. Live yeast supplementation in milk replacer led to a decrease of diarrhea in milk-fed veal calves and the fecal microbiota of diarrheic calves maintained a healthy community similar to nondiarrheic animals, with Fecalibacterium being the predominant genus. Key words: Saccharomyces cerevisiae boulardii, growth, fecal microbiota, diarrhea INTRODUCTION

High levels of mortality and morbidity before weaning in dairy replacement heifer and veal calves represent animal health, antimicrobial resistance, and welfare concerns, which in turn result in substantial economic losses for the industry due to operational costs and decreased production. In the milk-fed veal industry, the mortality rate has been recently found to be 7.6% in Canada (Winder et al., 2016) and 5.3% in Belgium, with 25.4% of the calves experiencing at least one disease between arrival and slaughter (Pardon et al., 2012). Although the mortality rate of dairy heifers in the United States decreased from 11% in 2007 (USDA, 2007) to 5% in 2014, the overall morbidity rate (33.9%) is still alarmingly high (Urie et al., 2018). In the studies previously cited, 2 common details were

7011

7012

VILLOT ET AL.

reported: (1) the main cause of death was digestive diseases and disorders, and (2) the highest morbidity incidence rate appeared during the first 4 wk of life. From birth, calves experience stressful situations, such as a new environment, fast diet transitions, exposure to infectious agents, and in the case of veal calves, transportation and commingling with calves from other environments (Bokma et al., 2019). With a naïve immune system, and a nonstandardized colostrum intake at birth, calf health, performance, and welfare can be adversely affected by digestive diseases and disorders (Barrington and Parish, 2001). Therefore, improving dairy calf health can play an important role in increasing production and profitability on the farm. In most cases, calves receive antimicrobial treatments in the first weeks of life to prevent or treat diseases. From a Dutch study, the average calf on a veal farm was exposed to at least one antimicrobial for 28.6 d/yr, which is significantly higher than pig exposure (9.3 d/yr; Bos et al., 2013). In the face of emerging antibiotic resistance threatening human health, restricting antibiotic use in livestock production has prompted research into potential alternatives to prevent diseases (McAllister et al., 2018). Probiotics are live microorganisms such as bacteria, yeast, and fungi that possess beneficial health effects for humans and animals. The probiotic Saccharomyces cerevisiae boulardii (SCB) has been shown in monogastric studies to decrease diarrhea and stimulate growth rates in pig and broiler production (Bontempo et al., 2006; Rajput et al., 2013; Hancox et al., 2015). In young calves, conflicting results have been highlighted concerning the effect of S. cerevisiae on health and performance (Alugongo et al., 2017). Supplementation of SCB showed no overall effects on performance, health, or fecal microbial groups in healthy calves (He et al., 2017). Nevertheless, for calves with failed passive transfer, a potential antidiarrheal effect of SCB supplementation could have helped to improve performance of the animal (Galvão et al., 2005). Similar results were also reported using other probiotics in young calves (Timmerman et al., 2005). For example, Oikonomou et al. (2013) reported lower bacterial diversity in the feces of calves with neonatal diarrhea and pneumonia, compared with healthy calves. Improving microbial diversity in the gut is one mode of action of SCB that has been characterized in monogastric experimental models to improve gut health (McFarland, 2010; Brousseau et al., 2015), and because neonatal and preweaning dairy calves function as monogastrics, probiotics such as SCB show strong promise. In the present study, we hypothesized that feeding the live yeast SCB in milk replacer (MR) could improve Journal of Dairy Science Vol. 102 No. 8, 2019

the health and performance of milk-fed veal calves by enriching beneficial gut microbiota, thus reducing the need of curative medical treatment. The objective of this study was to evaluate if dietary SCB could improve growth and performance and decrease calf mortality and morbidity. The study further aimed to explore if gut microbiota enrichment can be observed in feces samples of calves supplemented with SCB in milk. MATERIALS AND METHODS

This in vivo study was conducted in accordance with the Canadian Council on Animal Care Guidelines and Policies with approval from the Animal Care and Use Committee (AUP #00002015) for the University of Alberta and was carried out in a commercial milk-fed veal facility in Cambridge, Ontario, Canada. Calf Selection and Dietary Treatments

A total of 84 Holstein bull calves with an age of 6 ± 3 d (± SD) sourced from local Ontario dairy farms were transported to the veal facility on the same day (March 1, 2017) and were randomly assigned to a unique pen number that referred specifically to 1 of the 2 dietary treatments for 7 wk. Animals were individually examined for evidence of disease, injury, and dehydration, and those presenting an initial unhealthy status were not included in the total of 84 calves. Animals were then weighed and housed in individual (100 × 125 cm) pens with rubber flooring for the duration of the trial. This experiment did allow physical contact between different calves with different treatments but was minimized by placing 4 calves for each treatments beside one another. Eighty-four animals were required to determine a significant difference in performance between the 2 dietary treatments according to the power analysis performed by POWER procedure of SAS (version 9.0, SAS Institute Inc., Cary, NC). Calves were fed daily in plastic buckets in 2 equal volumes of MR at 0730 and 1730 h starting with a volume of 5 L/d (2.5 L/meal) during 7 d from their arrival and gradually increasing by 0.1 L/d to reach 15 L/d (7.5 L/meal) on wk 7 for each calf. The MR used in this study contained 260 g/kg of CP, 160 g/kg of crude fat, and 4.58 Mcal/kg of ME on a DM basis (Grober Animal Nutrition, Cambridge, Ontario, Canada), and 150 g of powder was mixed with water (42°C) to make 1 L of MR solution. A total of 42 calves were randomly assigned to the treated group (SCB) and were fed the commercial MR supplemented throughout the experiment with the same quantity of 5 g of live SCB product every afternoon feeding (ProTernative Milk; Levucell

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

SB20 containing the Pasteur Institute CNCM I-1079 strain of SCB; Lallemand Animal Nutrition, Montreal, Quebec, Canada), which was expected to supply 10 × 109 cfu of SCB per calf per day. The other 42 calves (CON) were fed with the same commercial MR, and a consistent 5 g/d of an inert material of powdered silica (Perkasil SM 660, Grace, Dueren, Germany) was added as a placebo during the PM feeding to maintain a blind experiment for the farm staff. Two students, blinded to the experimental group, were trained to be observers by a veterinary practitioner and were responsible for data collection. Week 0 of the experiment refers to the day that calves arrived at the facility, and wk 1 refers to the first 7 d after the day of arrival. No antibiotic treatments were administrated to the calves before their arrival. Water with electrolytes was provided on the day of arrival followed by free access to water for the duration of the experiment. From wk 0, the MR consumed made up the entire dietary intake of the calves. No metaphylactic treatment was given, but the daily monitoring of health parameters, behavior, and individual feed intake contributed to identifying an unhealthy animal. A decision tree (Supplemental Figure S1; https:​/​/​doi​.org/​10​.3168/​jds​.2018​-16149) was used to identify the disease beforehand and to determine necessary treatments for an individual calf. A specific intervention was followed strictly, first based on general appearance of the calves, then on fecal score, rectal temperature, umbilicus aspect, and nasal discharges. Calves with diarrhea received 2 L of oral electrolytes (Calf Lyte II, Vetoquinol, Lavaltrie, Quebec, Canada) twice a day: once between feedings and once in the evening. If no improvement was observed after electrolyte treatment or if severe diarrhea was detected, calves received trimethoprim sulfa (Trimidox, 0.7 mL/10 kg of BW, Vetoquinol) for 3 d. Ceftiofur (Excenel, 0.2 mL/10 kg of BW, Zoetis, Kirkland, Quebec, Canada) for 3 d was used as a second line therapy if the calf was still diarrheic. All calves with navel and respiratory problems were also treated and remained in the study for its duration. All mortalities during the experiment had a postmortem examination completed by a veterinarian to determine the cause of death. Measurements and Sample Collection

Milk replacer refusals were weighed and recorded daily during the experimental period to determine daily feed intakes. Calves were weighed at their arrival, the day after the start of the experiment, and weekly thereafter until wk 7, to determine weekly ADG. Calves were always weighed at the same time (1000 h) of day to ensure consistency. A jugular blood sample was col-

7013

lected at the arrival for the determination of IgG concentrations using radial immunodiffusion (Chelack et al., 1993) at the SCCL Ltd. (Saskatoon, SK, Canada) laboratory. Nose, joint, cough, navel, eye, and ear scores were recorded daily at 0930 h using a 0 to 3 scale developed by University of Wisconsin–Madison (McGuirk, 2008), where 0 is the healthy state. In addition, the standard health scoring system (Renaud et al., 2018a) was used for fecal, dehydration, and attitude scores. Briefly, fecal consistency was scored according to the following scale: 0 = normal; 1 = semi-formed, pasty; 2 = loose but stays on top of mat; and 3 = watery, sifts through the mat. Dehydration was based on skin turgor by pinching and stretching the skin on the neck and watching it rebound on both sides of the animal. The scale was 0 = skin tent returns to normal <2 s; 1 = skin tent returns to normal in 2 s; 2 = skin tent returns to normal in 2 to 4 s; and 3 = skin tent returns to normal in >4 s. The attitude score was based on whether the calf would get up when pinched on the spine at the hips, as well as the overall demeanor of looking dull or depressed and the suckling ability. The scale was 0 = gets up, perky and energetic, bright alert, and strong suckle; 1 = does not get up or looks slightly dull or depressed, eyes not sunken, and good suckle; 2 = will not get up, seems dull and depressed, eyes slightly sunken, and good suckle; 3 = will not even raise head and looks very dull and depressed, mild depression, sternal recumbency, moderately sunken eyes, and tacky mucus membranes with poor suckle. A diarrhea case was defined when the fecal score was ≥2 for at least 2 d (Lesmeister and Heinrichs, 2004). A severe diarrhea case was defined when dehydration and attitude scores >0 and fecal score ≥2. Duration of the illness, as well as the number of days necessary to recover (fecal score ≤1) after the initiation of diarrhea, were evaluated. Fecal samples were collected at arrival (wk 0), then again on d 14 (wk 2) and d 49 (wk 7), 3 h after the morning feeding of that day using sterile gloves. Fecal samples (20 g) were collected directly from the rectum by rectal palpation and samples were stored in 15-mL Falcon tubes on ice during sampling before being frozen at −20°C immediately after collection until further analysis. DNA Extraction from Fecal Samples

Total DNA from feces was extracted using the repeated bead beating plus column method (Yu and Morrison, 2004). Briefly, the sample (~0.2 ± 0.1 g) was washed twice with Tris-EDTA buffer. After the addition of cell lysis buffer containing 20% SDS, samples were subjected to physical disruption at 5,000 rpm for Journal of Dairy Science Vol. 102 No. 8, 2019

7014

VILLOT ET AL.

3 min using Biospec Mini Beads Beater 8 (BioSpec, Bartlesville, OK), followed by incubation at 70°C for 15 min and centrifugation for 5 min at 16,000 × g at 4°C. The bead-beating, incubation, and centrifugation were repeated once, and impurities were removed from the supernatant using 10 M ammonium acetate, followed by DNA precipitation using isopropanol. After precipitation, DNA was further purified using QIAmp fast DNA stool mini kit (Qiagen Inc., Germantown, MD). The DNA quantity and purity were evaluated using a NanoDrop 1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE) and stored at −20°C until further use. Estimation of Bacteria Groups and S. cerevisiae in Feces with Real-Time Quantitative PCR

The densities of total bacteria, Bifidobacterium, Lactobacillus, Fecalibacterium prausnitzii, and total E. coli were estimated by measuring their respective 16S rRNA gene copy numbers, and primer sets targeting the 26S rRNA gene of S. cerevisiae were used to estimate the copy numbers of SCB using real-time quantitative PCR (qPCR). The qPCR was performed in the highthroughput Viia 7 Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA) using SYBR green chemistry (Fast SYBR Green Master Mix, Applied Biosystems, Foster City, CA) with specific primers (Supplemental Table S1; https:​/​/​doi​.org/​10​.3168/​jds​ .2018​-16149). Standard curves were generated for each bacterial group using purified plasmid carrying 16S rRNA genes of Butyrivibrio hungatei, Bifidobacterium adolescentis, Lactobacillus acidophilus ATCC4356, Fecalibacterium prausnitzii A2–165, and Escherichia coli K12, respectively. The copy number of 16S rRNA per gram of fresh matter was calculated using the equation described by Li et al. (2009). The proportion of each group of bacteria was calculated as copy number per gram of each bacterial group divided by the density per gram of total bacteria (Malmuthuge et al., 2015). A genomic DNA of SCB was extracted and isolated directly from the ProTernative Milk containing CNCM I-1079 strain to generate the yeast standard curve to estimate the gene copy number of S. cerevisiae in calf fecal samples. Profiling of the Fecal Bacteria Using Amplicon Sequencing

The V1–V3 hypervariable region of the 16S rRNA gene for amplicon sequencing was amplified with 27F and 515R primers (27F-CS1F: 5′-ACACTGACGACATGGTTCTACAGAGTTTGATCMTGGCTCAG-3′, Journal of Dairy Science Vol. 102 No. 8, 2019

515R-CS2R: 5′-TACGGTAGCAGAGACTTGGTCTCCGCGGCKGCTGGCAC-3′; Kroes et al., 1999). The amplicon DNA with targeted size (~500 bp) was purified from 1% agarose gel using a QIAEX II gel extraction kit (Qiagen Inc.). The quality and quantity of purified PCR products were checked using a NanoDrop 1000 (NanoDrop Technologies, Wilmington, DE) to ensure that the concentration of DNA from all samples was higher than 25 ng/μL. The amplicons were sequenced at Genome Quebec at McGill University (Montreal, QC, Canada) using Illumina’s MiSeq (Illumina Inc., San Diego, CA) platform (2 × 250, pair-end). Taxonomic Identification of Fecal Bacteria

Sequence data were further analyzed using Quantitative Insight into Microbial Ecology (QIIME) package, version 1.9 (Caporaso et al., 2010). First, low-quality (Phred score <20) and short (<100 bp) reads were filtered out from the raw sequences. Then, the chimeric sequences were removed with usearch method (Edgar, 2010), and the remaining sequences were subjected to operational taxonomic unit (OTU) identification based on 97% similarity using the assign taxonomy function. Taxonomic characterization was performed using the SILVA database (SILVA Release 123, July 2015; https:​ / ​ / ​ w ww​. arb ​ - silva ​ . de/ ​ d ocumentation/ ​ r elease​ - 123/​ ) . Singleton OTU (represented by one read only) were removed from downstream analysis to reduce overprediction of rare OTU. Principal coordinate analysis of the microbial profiles was conducted using Bray-Curtis distance matrices in R (RStudio Version 1.1.456, Boston, MA). Statistical Analyses

The individual calf was considered the experimental unit. Data from calves that died were included in the data set and accounted for until date of death. Body weight, ADG, DMI, and milk intake (MI) were adjusted by the GLIMMIX procedure of SAS using initial BW as a covariate. Treatment, time (week), and their interaction were fixed effects with week being considered as a repeated measure. Means were calculated by the least significant difference method of SAS. The covariance structure for the data was tested and the structure that best fit the data (lowest Akaike information criterion and Bayesian information criterion) was chosen (Littell et al., 1998). Treatment differences with P ≤ 0.05 were considered significant, and 0.05 < P ≤ 0.10 was considered a tendency. When a treatment effect occurred, individual comparisons used the PDIFF adjusted = Bonferroni statement of SAS.

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

Incidence of mortality, diarrhea, and antibiotic treatments were analyzed with GLIMMIX procedure of SAS using a binomial distribution, and the predicted log of the odds of failure (i.e., mortality = 0) was modeled. The FREQ procedure was used to obtain proportions of calves within each group CON and SCB, and results were reported using percentage and number of animal in each group, odds ratio, and 95% confidence limits. Probability values presented are those relative to the odds ratio and 95% confidence intervals of logistic regression models. Data regarding fecal and dehydration scores, number of diarrhea cases, length of diarrhea, and the number of electrolyte and antibiotic treatments were analyzed with the same procedure but using a multinomial distribution. Data were categorized if necessary and the log odds of lower levels of the ordinal outcome were modeled. A total number of counts or average values for each group were reported. To determine if SCB supplementation affected diarrheic and nondiarrheic calves differentially, we defined calves that experienced at least one diarrhea case (fecal score ≥2 for at least 2 d) as diarrheic, whereas the others were qualified as nondiarrheic. The effect on performance data of dietary treatment (CON or SCB), diarrhea status (diarrheic or not), and their interaction was analyzed using the GLIMMIX procedure of SAS, as described previously. Differences of dietary treatment × diarrheic status interaction were considered significant with P ≤ 0.05 and as a trend when 0.05 < P ≤ 0.10. When an effect occurred, individual comparisons used the PDIFF adjusted = Bonferroni statement of SAS. The effects of SCB supplementation on the relative abundance of bacterial at phylum, family, and genus level, as well as the density of total bacteria, prevalence (% of total bacteria), and density of 4 bacterial species, were analyzed by Mann-Whitney U test at wk 2 and 7, respectively. Similar microbial analysis was also performed among nondiarrheic and diarrheic in CON and SCB groups at wk 2, respectively, by Kruskal-Wallis test. The effect of dietary treatment (CON or SCB) and diarrhea status and their interaction on microbial markers was analyzed using the aligned ranking transform test. A univariate permutation test was performed to determine bacterial genus differences among 4 groups. All microbial analysis was performed using R version 3.4.1. A significant difference was declared at P-value <0.05. Nucleotide Sequence Accession Numbers

All DNA sequences were deposited in the National Center for Biotechnology Information sequence read archive and are accessible under the accession numbers SRR8267761 to SRR8267982.

7015

RESULTS

Copy numbers of 26S rRNA S. cerevisiae gene detected in the feces of the calves at arrival were less than 2.2 × 105/g of fecal sample (Figure 1). Supplementation of SCB in milk significantly increased the density of S. cerevisiae per g of feces of the SCB calves at wk 2 (1.7 × 107) and wk 7 (1.2 × 107) compared with wk 0 (2.2 × 105; P < 0.05). The copy number of 26S rRNA S. cerevisiae gene in CON calves was not affected among wk 0, 2, and 7, with an average value of 1.5 × 105/g of feces (P = 0.312; Figure 1), indicating that it was unlikely that SCB cross-contamination to the CON calves occurred. Performance and Health Parameters of Calves

On arrival, calves were individually assessed for health parameters with a healthy attitude score, no sign of dehydration, an average body temperature of 39 ± 0.3°C, and a mean fecal score of 0.7 ± 0.8. Serum concentration of IgG at arrival was not different between the 2 dietary groups of calves (15.5 ± 8.5 and 16.8 ± 8.5 mg/mL, respectively, for CON and SCB, P = 0.468). Average daily gain followed similar patterns as DMI and BW with a slow increase from wk 1 to 2, or even a decrease of ADG at wk 2, followed by an approximately linear increase until wk 7 (wk: P < 0.001). The addition of SCB to the milk from wk 1 of age to wk 7 did not affect overall performance of the calves with an ADG of 585 ± 24 and 572 ± 22 g/d for CON and SCB calves, respectively (P = 0.702). The final BW was similar for both CON and SCB calves (74.5 ± 1.1 and 73.5 ± 1.0 kg, respectively, P = 0.491). The effects of milk supplemented with SCB on the health of calves are reported in Table 1. In total, 8 calves died during this experiment, including 6 calves in the CON group as a result of enteritis and colitis from wk 1 to 3 of the experiment. One calf supplemented with SCB died as a result of enteritis during wk 3 of the experiment, and the other one was euthanized at wk 2 due to a navel infection. Calves supplemented with SCB in milk had a lower odds of having a fecal score ≥2 in comparison with the CON calves (P < 0.001). The SCB supplementation specifically decreased the fecal score of the calves at wk 1, 2, and 4 in comparison with the CON group (Figure 2a; P < 0.05). During the entire experiment, 49 calves experienced diarrhea, with a larger proportion observed in the CON group (69.1% of diarrheic calves in the CON group and 50.0% in the SCB group). Adding SCB to the MR resulted in a tendency for decreased odds of experiencing diarrhea in comparison with the CON calves (P = 0.081) and significantly decreased odds of experiencing severe Journal of Dairy Science Vol. 102 No. 8, 2019

7016

VILLOT ET AL.

diarrhea for the SCB group (P = 0.036, Table 1). The number of days with scores ≥1 for attitude and dehydration parameters were higher in CON than SCB groups (P < 0.001). The number of calves treated with antibiotics to mitigate diarrhea was significantly lower in the SCB group compared with CON (14.3 vs. 33.3%, respectively, P = 0.048) and the average number of electrolytes used in SCB calves was reduced (5.1 vs. 6.4 for CON calves among the experiment, P = 0.024). Fecal score and diarrhea incidence were affected by week (P < 0.001); calves primarily (89.0%) experienced diarrhea during the first 2 wk after arrival (Figure 2b). During wk 1, when diarrhea incidence was the highest, diarrheic calves supplemented with SCB had lower odds to experience diarrhea (42.3 vs. 59.5% of diarrheic calves for SCB and CON groups, respectively). On average, 1.1 cases of diarrhea were reported for diarrheic calves supplemented with SCB during the experiment, whereas 1.4 cases of diarrhea were recorded for the CON calves (P = 0.112). No differences were found between the 2 groups regarding the number and the duration of diarrhea cases or the number of electrolytes used during the experiment (P > 0.100). The number of diarrheic calves treated with antibiotics tended to be lower when calves were supplemented with SCB (66.7%) compared with the CON calves (89.7%; P = 0.061). The number of days of antibiotic treatment was

similar for the 2 groups. Among the enrolled calves, in the CON group, the diarrheic calves had a lower ADG (524 ± 29 g/d) compared with the nondiarrheic calves (654 ± 38 g/d; P = 0.038), whereas in the SCB group, the diarrheic calves maintained the same ADG compared with the nondiarrheic calves (respectively 572 ± 32 and 572 ± 30 g/d; P > 0.100). The BW of the diarrheic CON calves (72.1 ± 1.3 kg) measured at wk 7 was lower compared with nondiarrheic calves (76.8 ± 1.6 kg; P = 0.048); however, diarrheic calves supplemented with SCB had a high BW (73.9 ± 1.4 kg) at wk 7, similar to the SCB nondiarrheic calves (73.0 ± 1.3 kg, P > 0.100; Table 2). Taxonomic Composition of Fecal Microbiota

A total of 8,523,893 high-quality reads were generated by amplicon sequencing with an average of 35,966 reads per sample assigned to 22,889 OTU (an average of 1,027 ± 38 OTU per sample) based on a 97% nucleotide sequence similarity. Firmicutes, Actinobacteria, Proteobacteria, Bacteroidetes, and Fusobacteria were predominant phylum in fecal samples at any sampling time for both groups (Figure 3). There was no effect of supplementation of SCB on the relative abundance of bacterial taxa at phylum, family, or genus levels at wk 0, 2, and 7. Furthermore, we compared the effect of

Figure 1. Changes in the copy number of 26S rRNA Saccharomyces cerevisiae gene per gram of fecal samples of calves fed with milk replacer supplemented with a placebo (CON) and calves fed with 5 g/d of S. cerevisiae boulardii CNCM I-1079 (SCB). Values are means ± SE. **P < 0.05 for comparison between CON and SCB groups with time and treatment and their interaction as fixed factors. TRT = treatment. Journal of Dairy Science Vol. 102 No. 8, 2019

7017

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

supplementation of SCB on the relative abundance of fecal bacterial taxa at the genus level between nondiarrheic and diarrheic calves at wk 2 when the incidence of diarrhea was the highest and when the ADG was negative. The relative abundance of the top 10 most abundant genera found in nondiarrheic and diarrheic calves from CON and SCB is presented in Table 3. Despite a slight difference in orders, the top 6 predominant genera in fecal samples of diarrheic calves supplemented with SCB were the same as those in nondiarrheic calves from both CON and SCB groups. Fecalibacterium was the most predominant bacterial genus in the fecal samples of nondiarrheic calves and in diarrheic calves supplemented with SCB. On the contrary, fecal microbiota was predominated by Collinsella in the diarrheic calves from the CON group, the relative abundance of which was 3 times higher than of Fecalibacterium. The univariate permutation test based on Bray-Curtis distance suggested the dissimilarity in bacterial genus composition at wk 2 among 4 groups (R2 = 0.07, P = 0.03; Supplemental Figure S2; https:​/​/​doi​.org/​10​.3168/​ jds​.2018​-16149). Specifically, the relative abundance of Fecalibacterium was lower in diarrheic calves in the CON group than nondiarrheic calves in both the CON

and SCB groups (P < 0.05), but not different between diarrheic calves supplemented with SCB and the other 3 groups (P > 0.05; Figure 4a). The relative abundance of Escherichia-Shigella (Figure 4b) tended to be higher in diarrheic calves compared with nondiarrheic calves in the CON group (P = 0.069), whereas no difference was measured in the SCB group between calves with different diarrheic status (P = 0.503). Shift in Fecal Microbial Groups

Further quantification of 4 bacterial groups using qPCR demonstrated that SCB supplementation and diarrhea status did not change the density of fecal total bacteria, Bifidobacterium, and E. coli (Table 4). The density of F. prausnitzii was significantly increased in diarrheic calves compared with nondiarrheic calves, whereas no change was reported in terms of the prevalence of F. prausnitzii among animal diarrheic status. Supplementation of SCB significantly increased (P = 0.01) the density of Lactobacillus in feces collected at wk 2 (3.85 × 105/g) compared with the CON group (11.55 × 105/g) and tended to increase its prevalence (P = 0.07) (Table 4). On the other hand, diarrheic

Table 1. Effect of Saccharomyces cerevisiae boulardii CNCM I-1079 supplementation on health parameters of calves during 7 wk of experiment Dietary treatment1 Health parameter Mortality [% (no.)] Diarrhea [% (no.)] Severe diarrhea [% (no.)] Calves treated with antibiotic [% (no.)] Calves treated with antibiotic specific to diarrhea [% (no.)] Average no. of days4 (% per calf)   Fecal score ≥2   Dehydration score ≥1   Attitude score ≥1   Eye score ≥1   Nose score ≥1   Ear score ≥1   Cough score ≥1   Joint score ≥1   Navel score ≥1   Average no. of electrolytes used, per calf Calves experiencing at least 1 diarrhea case   Calves treated with antibiotic [% (no.)]   Average no. of diarrhea cases, per calf   Average length of diarrhea (no. of days per calf)   Average no. of electrolytes used, per calf

CON 14.3 69.1 28.6 81.0 33.3 9.7 10.1 5.9 43.8 8.0 18.9 18.6 1.1 12.5 6.4

(6/42) (29/42) (12/42) (34/42) (13/42)  

  89.7 (26/29) 1.4 2.8 6.9

SCB 4.8 50.0 9.5 66.7 14.3 5.9 5.4 3.4 46.1 9.0 23.3 19.6 1.1 9.8 5.1

(2/42) (21/42) (4/42) (28/42) (6/42)  

  66.7 (14/21) 1.1 2.8 5.3

OR2

95% CI

P-value3

3.33 2.23 3.80 2.13 3.00   1.50 2.17 1.91 0.91 0.88 0.78 0.93 1.00 1.45 5.63   4.33 3.29 1.66 1.58

0.62–18.0 0.90–5.51 1.10–13.23 0.77–5.88 1.00–8.95   1.18–1.90 1.71–2.75 1.42–2.58 0.93–1.20 0.77–1.19 0.73–0.98 0.89–1.22 0.61–1.84 1.19–1.76 1.25–25.30   0.93–20.20 0.75–14.40 0.53–5.17 0.54–4.67

0.160 0.081 0.036 0.144 0.048   <0.001 <0.001 <0.001 0.347 0.684 0.031 0.610 0.780 0.101 0.024   0.061 0.112 0.372 0.399

1 CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of S. cerevisiae boulardii CNCM I-1079. 2 OR = odds ratio referring to CON. For binary variable, the predicted log of the odds of failure has been modeled (i.e., mortality = 0). For multinomial variables, the log odds of lower levels of the ordinal outcome, in the sequence encountered in the data, has been modeled (i.e., lower number of diarrhea cases). 3 P-value of comparison between CON and SCB groups relative to logistic regression models, OR, and 95% CI. 4 Based on 50 d of recorded scores per animal or until death.

Journal of Dairy Science Vol. 102 No. 8, 2019

7018

VILLOT ET AL.

Figure 2. Effect of Saccharomyces cerevisiae boulardii CNCM I-1079 supplementation on the average fecal score (a) and total diarrhea cases (b) of the calves (initial n = 82) among time. CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of S. cerevisiae boulardii CNCM I-1079. Values are means + SE with initial BW used as a covariate. Statistics were performed with treatment (TRT), time (week), and their interaction as fixed effects with week being considered as a repeated measure. **Adjusted P < 0.05 for the comparison between CON and SCB among weeks.

Journal of Dairy Science Vol. 102 No. 8, 2019

7019

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

calves in CON group had a lower prevalence of fecal Lactobacillus than those nondiarrheic calves in the CON group at wk 2 (P < 0.05). DISCUSSION

Probiotic supplementation in MR could help to mitigate the adverse effects of milk-fed veal industry, especially when animals experience diarrhea. In this study, the effects of administering daily SCB CNCM I-1079, at arrival to the farm to wk 7, on performance, health, and fecal microbiota of calves were investigated. Successful Establishment of SCB in Milk-Fed Calves

The S. cerevisiae was quantified in the fecal content of calves receiving daily doses of SCB. With the same quantity of ProTernative Milk distributed to each calf (10 × 109 cfu), the recoveries in fecal samples were 1.7 × 107 and 1.2 × 107 copy numbers/g of fecal matter at wk 2 and 7, respectively. In a study where the ProTernative Milk supplementation was 7.5 × 108 cfu/L of MR + 3 × 109 cfu/kg of starter feed, the authors confirmed additionally the viability of the yeast in fecal samples by the enumeration of viable cell counts (Fomenky et al., 2017). The SCB recovered in the order of 5.0 × 105 cfu/mL throughout the gut and in fecal samples suggests that the yeast survives in significant amounts in the digestive tracts of calves. In the CON group, an insignificant quantity of S. cerevisiae DNA was detected. Those results confirm the successful establishment of SCB in the treated calves during the experiment and the absence of significant contaminations that could occur with potential wild S. cerevisiae yeast exposure (Urubschurov et al., 2008).

Growth of Diarrheic Milk-Fed Calves Maintained with SCB

The absence of growth and performance improvement in calves supplemented with SCB was demonstrated previously during the preweaning period (He et al., 2017), as well as postweaning (Fomenky et al., 2017). The authors suggest that the beneficial effects of SCB supplementation could be improved when environmental conditions lead to high disease challenges. In our study, the mortality rate (9.5%) was higher than previously reported in Ontario veal farms (Winder et al., 2016). The lack of metaphylactic treatment provided on arrival, or at any time during observation, combined with the high diarrhea incidence of CON calves during the first weeks of the experiment, may have resulted in the observed high mortality. Failure of passive transfer has been associated with early mortality (Renaud et al., 2018b). When we focused our analysis by differencing calves who experienced diarrhea or not within each dietary treatment, we found that the diarrheic calves supplemented with SCB maintained the same ADG, MI, and final BW in comparison to the nondiarrheic calves, whereas the CON diarrheic calves had significant lower growth parameters in comparison to the CON nondiarrheic calves. In clinical trials, lower weight gain is reported when sick calves are treated with antibiotics in comparison to healthy and untreated calves (Berge et al., 2005); however, in the SCB group of our study, calves experiencing diarrhea and treated with an antibiotic could still maintain similar growth as nondiarrheic calves. These findings are in accordance with the effect of live SCB dietary supplementation on improved growth performance in piglets with stress during weaning (Bontempo et al., 2006).

Table 2. Average growth parameters of the calves according to their diarrheic status and dietary treatments1,2 Group CON Item ADG (g/d) Final BW (wk 7; kg) Feed efficiency3 Milk intake (L/d)

No-DC (n = 13) 654 76.8 0.44 9.0

± ± ± ±

38a 1.6a 0.05 0.2a

SCB DC (n = 29)

524 72.1 0.28 8.3

± ± ± ±

29b 1.3b 0.03 0.2b

         

No-DC (n = 21) 572 73.0 0.39 8.8

± ± ± ±

30ab 1.3ab 0.04 0.2ab

P-value DC (n = 21) 572 73.9 0.34 8.7

± ± ± ±

32ab 1.4ab 0.04 0.2ab

 

TRT

DC status

Group (TRT × DC)

       

0.702 0.491 0.905 0.796

0.035 0.166 0.014 0.042

0.038 0.050 0.190 0.073

a,b

Different superscripts indicate a significant difference among DC status and dietary treatment interaction (P < 0.05). Values are LSM ± SE with initial BW used as a covariate. n = number of animals per group before death. 2 CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of Saccharomyces cerevisiae boulardii CNCM I-1079; DC = diarrheic calves; no-DC = nondiarrheic calves; TRT = dietary treatment. 3 Feed efficiency = milk intake (g/d)/ADG (g/d). 1

Journal of Dairy Science Vol. 102 No. 8, 2019

7020

VILLOT ET AL.

Health Parameters of Milk-Fed Calves Improved with SCB

No direct measurement of intestinal health was performed during this experiment, but daily monitoring of fecal scores and treatments allowed for a good determination of the health status of each animal individually. The health status according to the attitude, fecal, and dehydration scores of the animals was more challenged during the first 2 wk of the experiment, with high incidence of disease and treatments and low ADG. This pattern is consistent with previous studies in veal calves, with the highest incidence of disease occurring in the first 3 wk following arrival (Pardon et al., 2012; Winder et al., 2016). Based on both the reduction of severe diarrhea incidence, and the maintained performance of diarrheic calves, SCB supplementation could help the animals to overcome the challenging period of early life in the milk-fed veal calf industry where birth management conditions are often unknown. The beneficial effects of SCB supplementation on fecal score and diarrhea have been also reported during the neonatal

period of piglets. In fact, in a group of piglets that was housed from birth with a litter treated by an individual dose of paste containing 3.3 × 109 cfu of dried SCB (Levucell SB, S. cerevisiae CNCM I-1079; Lallemand SAS, Blagnac, France), the authors reported lower fecal scores and fewer numbers of diarrheic days during the first week of life (Hancox et al., 2015). From an economic point of view, the lower mortality and morbidity for the calves supplemented with SCB decreased operational costs in this group for the veal facility, despite the lack of evidence supporting SCB as a growth promoter. The initially higher operational cost of the calves supplemented with SCB was surpassed in the CON group due to the more abundant use of health treatments. Fecal Microbial Composition of Diarrheic Calves Supplemented with SCB

Although the effect of probiotics on gut microbial composition has been extensively studied in swine and poultry (Ma et al., 2018), it is still largely unknown

Figure 3. Taxonomy analysis of 5 bacterial phyla detected (the relative abundance >0.1% and presented in more than half of the samples per group) in fecal samples of calves in CON and SCB groups at each sampling time. CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of Saccharomyces cerevisiae boulardii CNCM I-1079. Journal of Dairy Science Vol. 102 No. 8, 2019

7021

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

how probiotics potentially affect the gut microbiota in calves. In the current study, fecal samples were used to conduct microbial analysis, as they are noninvasive and easy to collect repeatedly. Failure to detect a change in the relative abundance of fecal bacterial genus between CON and SCB calves suggests that the effect of SCB in reducing the rate of diarrhea is not necessarily exhibited by changing fecal microbiota. In addition, Fomenky et al. (2018) observed significant changes in gut mucosa/ digesta microbial composition in calves supplemented with SCB before and after weaning. Notably, authors observed an increased relative abundance of beneficial bacterial genera, such as Ruminococcaceae UCG 005, Roseburia, and Olsenella, which might be related to diarrhea. Therefore, intestinal mucosa samples should be collected in future studies with a higher number of animals, as mucosa-associated bacteria interact more closely with the host gut health (Van den Abbeele et al., 2011). The potential modulation of host immune function, which can be mediated through regulating the gut microbiota, could be therefore highlighted. However, it should be noted that only 4 calves per treatment were used for microbial analysis in their experimental study and the small sample size might overestimate the treatment effect (Sterne et al., 2000). Since the difference in structure and membership of fecal microbiota between healthy and diarrheic calves has already been proven (Gomez et al., 2017), we further

investigated if any differences were present in microbial compositions in the groups based on the interaction between dietary treatments and diarrheic status. The use of fecal samples collected within 10 d after a detected diarrhea case is often reported in most of the studies to highlight a significant effect (Oikonomou et al., 2013; Gomez et al., 2017). Since the incidence of diarrhea was the highest between arrival to the farm and wk 2, we therefore used the wk 2 time point to investigate the potential perturbation of fecal microbiota due to diarrhea. In contrast to the diarrheic calves from the SCB group, we observed that the most predominant bacterial genus as well as several microbial markers in fecal samples of diarrheic calves from the CON group were different from those of nondiarrheic calves from both CON and SCB group. Fecalibacterium is known as a butyrate producer, and butyrate may enhance the integrity of the intestinal epithelial barrier (Schwiertz et al., 2010). A higher prevalence of Fecalibacterium spp. in fecal samples of newborn calves was associated with higher weight gain and lower diarrhea incidence in preweaning dairy calves (Oikonomou et al., 2013), whereas a lower prevalence of Fecalibacterium spp. was found to be decreased in dogs with acute diarrhea (Suchodolski et al., 2012). On the other hand, Collinsella can increase gut permeability by lowering the expression of tight junction protein in intestinal epithelial cells in vitro, and also enhance the expression of

Table 3. The relative abundance (median and ranges expressed as %) of the 10 most abundant genera from fecal samples of nondiarrheic and diarrheic calves in response to the supplementation of Saccharomyces cerevisiae boulardii CNCM I-1079 at wk 2 Group1 CON

SCB

No-DC (n = 13)

DC (n = 25)

 

No-DC (n = 20)

DC (n = 21)

Fecalibacterium 16.4 (2.3–65.9) Collinsella 3.8 (1.2–49.8) Blautia 3.0 (0.8–8.9) Lachnospiraceae UCG-008 2.3 (0.5–10.4) Subdoligranulum 2.1 (0.1–13.7) R. torques 1.7 (0.6–10.6) Coprococcus 1.3 (0.2–5.5) Lachnospiraceae NC2004 1.1 (0.1–7.3) Bacteroides 0.5 (0–21.7) E. rectale 0.4 (0–1.5)

Collinsella 9.8 (1.2–39.5) Blautia 3.4 (0–12.9) Fecalibacterium 3.0 (0.1–68.3) Lachnospiraceae UCG-008 2.6 (0–14.2) Ruminococcus torques 2.1 (0.5–11.0) Dorea 1.5 (0–11.1) Lachnoclostridium 1.0 (0.1–6.0) Eubacterium rectale 0.7 (0–2.8) Streptococcus 0.6 (0–46.3) Butyricicoccus 0.5 (0–27.9)

                                       

Fecalibacterium 21.9 (0.1–70.8) Collinsella 5.9 (0.2–33.2) Blautia 4.2 (0.1–16.4) Subdoligranulum 2.5 (0–18.2) Lachnospiraceae UCG-008 2.4 (0.2–21.3) R. torques 1.6 (0.1–4.5) Bacteroides 1.3 (0–21.6) Dorea 1.0 (0–9.3) Lachnoclostridium 0.9 (0.1–4.8) Coprococcus 0.7 (0–6.1)

Fecalibacterium 12.1 (0–77.8) Collinsella 5.4 (0.1–72.0) Lachnospiraceae UCG-008 1.5 (0–9.7) Blautia 1.5 (0–9.2) Subdoligranulum 1.2 (0–6.7) R. torques 0.9 (0.2–6.4) Bacteroides 0.8 (0–45.9) Alloprevotella 0.7 (0–12.7) Lachnoclostridium 0.7 (0–2.8) Dorea 0.7 (0–5.2)

1 CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of S. cerevisiae boulardii CNCM I-1079; DC = diarrheic calves; no-DC = nondiarrheic calves; n = number of animal sampled by group.

Journal of Dairy Science Vol. 102 No. 8, 2019

7022

VILLOT ET AL.

pro-inflammatory factors such as IL-17, both of which contribute to the loss of gut epithelial integrity (Chen et al., 2016). Lactobacillus has been proven to reduce the incidence of diarrhea in calves (Nagashima et al., 2010), whereas Escherichia-Shigella includes several opportunistic pathogens such as enteropathogenic E. coli and Shiga toxin–producing E. coli (Escobar-Páramo et al., 2004) that can cause diarrhea in calves. An increase in the abundance of Lactobacillus and decrease of E. coli could help ensure normal growth performance of calves even when they experience diarrhea (Abu-Tarboush et al., 1996). Individual antibiotic treatment was unavoidable in this study performed on a commercial farm to ensure the profitability of calf production when the health of animal was compromised; overall, this could have affected the fecal microbiota data. As part of a cohort study, Oultram et al. (2015) presented differences in the prevalence of Lactobacillus genus and increased species richness in therapeutic antibiotic-treated calves (n = 7) compared with controls (n = 7) by analyzing fecal microbiota 1 wk post-treatment. Nevertheless, the Chao1 indices of diversity 2 wk post-treatment were identical between the 2 groups. In addition, contrary to their expectations, therapeutic level of oxytetracycline administrated to 14 calves (1 g in 2 L of MR twice a day for 5 d) did not result in a shift of fecal community structure nor in species diversity compared with untreated calves (Keijser et al., 2019). Therefore, it seems that the administration of individual therapeutic treatment, also related to differences in health status between the animals, could be negligible compared with the potential effect of diarrhea on fecal community structure. These results suggest that a healthy gut microbial composition and membership in diarrheic calves was preserved in the calves supplemented with SCB, which might be important to maintain gut health, as well as similar growth and performance as those nondiarrheic calves in the current study. Future studies are needed to illustrate the interaction between gut microbiota and host immune function in response to the supplementation of SCB, as well as how these changes contribute to the reduced incidence of diarrhea or improved growth performance and health status in calves. Figure 4. Relative abundance (%) of (a) Fecalibacterium and (b) Escherichia-Shigella in fecal samples of nondiarrheic and diarrheic calves in response to the supplementation of SCB at wk 2. Different letters indicate a significant difference (P < 0.05). CON = calves fed with milk replacer supplemented with placebo. SCB = calves fed with milk replacer supplemented with 5 g/d of Saccharomyces cerevisiae boulardii CNCM I-1079; DC = diarrheic calves no-DC = nondiarrheic calves. The three horizontal lines of each box represent the first, second (median), and third quartiles, respectively, with the whiskers extending to 1.5 interquartile range. The black dots indicate outliers. Journal of Dairy Science Vol. 102 No. 8, 2019

CONCLUSIONS

In this study, we demonstrated that SCB has beneficial effects on health parameters of milk-fed calves raised in a commercial veal farm environment. Fewer calves experienced severe diarrhea in the SCB group and a reduction occurred in the therapeutic treatments used. In addition, the diarrheic calves supplemented

1.9 (0.2–19.7)   4.7ab (0–35.9) 0.0016ab (0.0001–0.065)   1.7 (0–44.9) 0.075 (0.0001–0.45)   4.0 (0.06–39.4) 1.29 (0.031–16.0)   2.2 (0.1–10.1) 0.0051 (0.00077–0.21)

No-DC (n = 13) 2.1 (0.2–16.8)   3.0b (0–76.1) 0.0006b (0–0.65)   3.4 (0–46.2) 0.055 (0.0003–1.87)   2.5 (0–24.8) 1.56 (0.0038–8.67)   2.3 (0.04–366) 0.0093 (0.00019–1.91)

DC (n = 25)

     

     

     

       

  1.9 (0.1–14.0)   11.8a (0–3640) 0.0094 (0.0007–6.80)a   0.7 (0–23.7) 0.032 (0.00004–0.27)   2.7 (0–22.3) 0.68 (0–8.68)   1.1 (0.1–39.6) 0.0058 (0.0005–0.32)

No-DC (n = 20)

SCB

3.4 (0.6–25.9)   11.3ab (0–1530) 0.002 (0.0001–0.48)ab   2.8 (0–148) 0.066 (0.0003–3.54)   4.4 (0–52.4) 1.89 (0.0078–11.0)   1.6 (0.08–54.8) 0.0064 (0.0008–0.13)

DC (n = 21)

     

     

     

       

 

  0.74 0.44

  0.52 0.88

  0.51 0.35

0.69   0.01 0.07

TRT

  0.43 0.27

  0.44 0.69

  0.02 0.22

0.13   0.15 0.64

DC status

P-value

  0.53 0.18

  0.14 0.12

  0.59 0.98

0.73   0.30 0.11

Group (TRT × DC)

1

Different superscripts indicate a significant difference among DC status and dietary treatment interaction (P < 0.05). Values of density and proportion of markers [(density of marker/density of total bacteria) × 100] are reported as median and ranges per gram of fresh samples. 2 CON = calves fed with milk replacer supplemented with placebo; SCB = calves fed with milk replacer supplemented with 5 g/d of S. cerevisiae boulardii CNCM I-1079; DC = diarrheic calves; no-DC = nondiarrheic calves; TRT = dietary treatment; n = number of animals sampled by group.

a,b

Escherichia coli   Density (107/g)   Proportion (%)

Bifidobacterium   Density (109/g)   Proportion (%)

Faecalibacterium prausnitzii   Density (108/g)   Proportion (%)

Total bacteria (1011/g) Lactobacillus   Density (105/g)   Proportion (10−1 %)

Items

CON

Group

Table 4. Density of total bacteria and 4 bacterial markers in fecal samples of nondiarrheic and diarrheic calves in response to the supplementation of Saccharomyces cerevisiae boulardii CNCM I-1079 at wk 21,2 YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

7023

Journal of Dairy Science Vol. 102 No. 8, 2019

7024

VILLOT ET AL.

with SCB maintained their growth performance, which might be explained by a similar fecal microbial composition in diarrheic calves supplemented with SCB as those nondiarrheic calves in the current study. With the increasing concern of antibiotic resistance occurring in animal production, SCB might also play a role in the reduction of treatments by stimulating improved intestinal health and reducing the occurrence of diarrhea. Fungal probiotics have the additional benefit of being resistant to antibiotics (Czerucka et al., 2007). Further studies should be performed to better understand the effect of SCB on gut microbiota by directly analyzing gut content or mucosa and exploring other potential modes of action of the yeast on the immune system and intestinal integrity. ACKNOWLEDGMENTS

This study was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Lallemand Animal Nutrition, Grober Animal Nutrition, Westgen, BC Dairy Association, Alberta Milk, SaskMilk, and Dairy Farmers of Manitoba and by Chinese Scholarship Council (CSC) Scholarship. We are grateful to Frédérique Chaucheyras-Durand, Mathieu Castex, and Angel Aguilar (Lallemand Animal Nutrition) and Amanda Kerr and Heather Copland (Grober Animal Nutrition) for their technical support and guidance. REFERENCES Abu-Tarboush, H. M., M. Y. Al-Saiady, and A. H. K. El-Din. 1996. Evaluation of diet containing lactobacilli on performance, fecal coliform, and lactobacilli of young dairy calves. Anim. Feed Sci. Technol. 57:39–49. https:​/​/​doi​.org/​10​.1016/​0377​-8401(95)00850​-0. Alugongo, G. M., J. Xiao, Z. Wu, S. Li, Y. Wang, and Z. Cao. 2017. Utilization of yeast of Saccharomyces cerevisiae origin in artificially raised calves. J. Anim. Sci. Biotechnol. 8:34. https:​/​/​doi​.org/​10​ .1186/​s40104​-017​-0165​-5. Barrington, G. M., and S. M. Parish. 2001. Bovine neonatal immunology. Veterinary Clinics: Food Animal Practice 17:463–476. Berge, A., P. Lindeque, D. Moore, and W. Sischo. 2005. A clinical trial evaluating prophylactic and therapeutic antibiotic use on health and performance of preweaned calves. J. Dairy Sci. 88:2166–2177. Bokma, J., R. Boone, P. Deprez, and B. Pardon. 2019. Risk factors for antimicrobial use in veal calves and the association with mortality. J. Dairy Sci. https:​/​/​doi​.org/​10​.3168/​jds​.2018​-15211. Bontempo, V., A. Di Giancamillo, G. Savoini, V. Dell’Orto, and C. Domeneghini. 2006. Live yeast dietary supplementation acts upon intestinal morpho-functional aspects and growth in weanling piglets. Anim. Feed Sci. Technol. 129:224–236. Bos, M. E., F. J. Taverne, I. M. van Geijlswijk, J. W. Mouton, D. J. Mevius, and D. J. Heederik. 2013. Consumption of antimicrobials in pigs, veal calves, and broilers in the Netherlands: Quantitative results of nationwide collection of data in 2011. PLoS One 8:e77525. https:​/​/​doi​.org/​10​.1371/​journal​.pone​.0077525. Brousseau, J.-P., G. Talbot, F. Beaudoin, K. Lauzon, D. Roy, and M. Lessard. 2015. Effects of probiotics Pediococcus acidilactici strain MA18/5M and Saccharomyces cerevisiae ssp. boulardii strain SBCNCM I-1079 on fecal and intestinal microbiota of nursing and Journal of Dairy Science Vol. 102 No. 8, 2019

weanling piglets. J. Anim. Sci. 93:5313–5326. https:​/​/​doi​.org/​10​ .2527/​jas​.2015​-9190. Caporaso, J. G., J. Kuczynski, J. Stombaugh, K. Bittinger, F. D. Bushman, E. K. Costello, N. Fierer, A. G. Pena, J. K. Goodrich, and J. I. Gordon. 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7:335. https:​/​/​doi​.org/​ 10​.1038/​nmeth​.f​.303. Chelack, B. J., P. S. Morley, and D. M. Haines. 1993. Evaluation of methods for dehydration of bovine colostrum for total replacement of normal colostrum in calves. Can. Vet. J. 34:407. Chen, J., K. Wright, J. M. Davis, P. Jeraldo, E. V. Marietta, J. Murray, H. Nelson, E. L. Matteson, and V. Taneja. 2016. An expansion of rare lineage intestinal microbes characterizes rheumatoid arthritis. Genome Med. 8:43. https:​/​/​doi​.org/​10​.1186/​s13073​-016​ -0299​-7. Czerucka, D., T. Piche, and P. Rampal. 2007. Review article: Yeast as probiotics - Saccharomyces boulardii. Aliment. Pharmacol. Ther. 26:767–778. https:​/​/​doi​.org/​10​.1111/​j​.1365​-2036​.2007​.03442​.x. Edgar, R. C. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–2461. Escobar-Páramo, P., K. Grenet, A. Le Menac’h, L. Rode, E. Salgado, C. Amorin, S. Gouriou, B. Picard, M. C. Rahimy, and A. Andremont. 2004. Large-scale population structure of human commensal Escherichia coli isolates. Appl. Environ. Microbiol. 70:5698–5700. Fomenky, B. E., J. Chiquette, N. Bissonnette, G. Talbot, P. Y. Chouinard, and E. M. Ibeagha-Awemu. 2017. Impact of Saccharomyces cerevisiae boulardii CNCMI-1079 and Lactobacillus acidophilus BT1386 on total lactobacilli population in the gastrointestinal tract and colon histomorphology of Holstein dairy calves. Anim. Feed Sci. Technol. 234:151–161. https:​/​/​doi​.org/​10​.1016/​j​ .anifeedsci​.2017​.08​.019. Fomenky, B. E., D. N. Do, G. Talbot, J. Chiquette, N. Bissonnette, Y. P. Chouinard, M. Lessard, and E. M. Ibeagha-Awemu. 2018. Direct-fed microbial supplementation influences the bacteria community composition of the gastrointestinal tract of pre-and postweaned calves. Sci. Rep. 8:14147. https:​/​/​doi​.org/​10​.1038/​s41598​ -018​-32375​-5. Galvão, K. N., J. E. Santos, A. Coscioni, M. Villaseñor, W. M. Sischo, and A. C. B. Berge. 2005. Effect of feeding live yeast products to calves with failure of passive transfer on performance and patterns of antibiotic resistance in fecal Escherichia coli. Reprod. Nutr. Dev. 45:427–440. https:​/​/​doi​.org/​10​.1051/​rnd:​2005040. Gomez, D. E., L. Arroyo, M. Costa, L. Viel, and J. Weese. 2017. Characterization of the fecal bacterial microbiota of healthy and diarrheic dairy calves. J. Vet. Intern. Med. 31:928–939. https:​/​/​doi​ .org/​10​.1111/​jvim​.14695. Hancox, L. R., M. Le Bon, P. Richards, D. Guillou, C. E. Dodd, and K. Mellits. 2015. Effect of a single dose of Saccharomyces cerevisiae var. boulardii on the occurrence of porcine neonatal diarrhoea. Animal 9:1756–1759. https:​/​/​doi​.org/​10​.1017/​S1751731114002687. He, Z., B. Ferlisi, E. Eckert, H. Brown, A. Aguilar, and M. Steele. 2017. Supplementing a yeast probiotic to pre-weaning Holstein calves: Feed intake, growth and fecal biomarkers of gut health. Anim. Feed Sci. Technol. 226:81–87. https:​/​/​doi​.org/​10​.1016/​j​ .anifeedsci​.2017​.02​.010. Keijser, B. J., V. Agamennone, T. J. van den Broek, M. Caspers, A. van de Braak, R. Bomers, M. Havekes, E. Schoen, M. van Baak, and D. Mioch. 2019. Dose-dependent impact of oxytetracycline on the veal calf microbiome and resistome. BMC Genomics 20:65. Kroes, I., P. W. Lepp, and D. A. Relman. 1999. Bacterial diversity within the human subgingival crevice. Proc. Natl. Acad. Sci. USA 96:14547–14552. Lesmeister, K., and A. Heinrichs. 2004. Effects of corn processing on growth characteristics, rumen development, and rumen parameters in neonatal dairy calves. J. Dairy Sci. 87:3439–3450. https:​/​/​ doi​.org/​10​.3168/​jds​.S0022​-0302(04)73479​-7. Li, X.-R., B. Du, H.-X. Fu, R.-F. Wang, J.-H. Shi, Y. Wang, M. S. Jetten, and Z.-X. Quan. 2009. The bacterial diversity in an anaerobic ammonium-oxidizing (anammox) reactor community. Syst. Appl. Microbiol. 32:278–289. https:​/​/​doi​.org/​10​.1016/​j​.syapm​ .2009​.03​.002.

YEAST SUPPLEMENTATION FOR MILK-FED VEAL CALVES

Littell, R., P. Henry, and C. Ammerman. 1998. Statistical analysis of repeated measures data using SAS procedures. J. Anim. Sci. 76:1216–1231. Ma, T., L. L. Guan, and Y. Suzuki. 2018. Dissect the mode of action of probiotics in affecting host-microbial interactions and immunity in food producing animals. Vet. Immunol. Immunopathol. https:​/​ /​doi​.org/​10​.1016/​j​.vetimm​.2018​.10​.004. Malmuthuge, N., Y. Chen, G. Liang, and L. A. Goonewardene. 2015. Heat-treated colostrum feeding promotes beneficial bacteria colonization in the small intestine of neonatal calves. J. Dairy Sci. 98:8044–8053. https:​/​/​doi​.org/​10​.3168/​jds​.2015​-9607. McAllister, T. A., Y. Wang, M. S. Diarra, T. Alexander, and K. Stanford. 2018. Challenges of a one-health approach to the development of alternatives to antibiotics. Anim. Front. 8:10–20. https:​/​/​ doi​.org/​10​.1093/​af/​vfy002. McFarland, L. V. 2010. Systematic review and meta-analysis of Saccharomyces boulardii in adult patients. World J. Gastroenterol. 16:2202. McGuirk, S. M. 2008. Disease management of dairy calves and heifers. Vet. Clin. North Am. Food Anim. Pract. 24:139–153. https:​/​/​doi​ .org/​10​.1016/​j​.cvfa​.2007​.10​.003. Nagashima, K., D. Yasokawa, K. Abe, R. Nakagawa, T. Kitamura, T. Miura, and S. Kogawa. 2010. Effect of a Lactobacillus species on incidence of diarrhea in calves and change of the microflora associated with growth. Biosci. Microflora 29:97–110. https:​/​/​doi​.org/​10​ .12938/​bifidus​.29​.97. Oikonomou, G., A. G. V. Teixeira, C. Foditsch, M. L. Bicalho, V. S. Machado, and R. C. Bicalho. 2013. Fecal microbial diversity in pre-weaned dairy calves as described by pyrosequencing of metagenomic 16S rDNA. Associations of Faecalibacterium species with health and growth. PLoS One 8:e63157. https:​/​/​doi​.org/​10​.1371/​ journal​.pone​.0063157. Oultram, J., E. Phipps, A. Teixeira, C. Foditsch, M. Bicalho, V. Machado, R. Bicalho, and G. Oikonomou. 2015. Effects of antibiotics (oxytetracycline, florfenicol or tulathromycin) on neonatal calves’ faecal microbial diversity. Vet. Rec. 177:598. Pardon, B., K. De Bleecker, M. Hostens, J. Callens, J. Dewulf, and P. Deprez. 2012. Longitudinal study on morbidity and mortality in white veal calves in Belgium. BMC Vet. Res. 8:26. https:​/​/​doi​.org/​ 10​.1186/​1746​-6148​-8​-26. Rajput, I., L. Li, X. Xin, B. Wu, Z. Juan, Z. Cui, D. Yu, and W. Li. 2013. Effect of Saccharomyces boulardii and Bacillus subtilis B10 on intestinal ultrastructure modulation and mucosal immunity development mechanism in broiler chickens. Poult. Sci. 92:956–965. https:​/​/​doi​.org/​10​.3382/​ps​.2012​-02845. Renaud, D. L., T. Duffield, S. LeBlanc, S. Ferguson, D. Haley, and D. Kelton. 2018a. Risk factors associated with mortality at a milkfed veal calf facility: A prospective cohort study. J. Dairy Sci. 101:2659–2668.

7025

Renaud, D. L., T. Duffield, S. LeBlanc, D. Haley, and D. Kelton. 2018b. Clinical and metabolic indicators associated with early mortality at a milk-fed veal facility: A prospective case-control study. J. Dairy Sci. 101:2669–2678. https:​/​/​doi​.org/​10​.3168/​jds​ .2017​-14042. Schwiertz, A., M. Jacobi, J.-S. Frick, M. Richter, K. Rusch, and H. Köhler. 2010. Microbiota in pediatric inflammatory bowel disease. J. Pediatr. 157:240–244. https:​/​/​doi​.org/​10​.1016/​j​.jpeds​.2010​.02​ .046. Sterne, J. A., D. Gavaghan, and M. Egger. 2000. Publication and related bias in meta-analysis: Power of statistical tests and prevalence in the literature. J. Clin. Epidemiol. 53:1119–1129. Suchodolski, J. S., S. E. Dowd, V. Wilke, J. M. Steiner, and A. E. Jergens. 2012. 16S rRNA gene pyrosequencing reveals bacterial dysbiosis in the duodenum of dogs with idiopathic inflammatory bowel disease. PLoS One 7:e39333. https:​/​/​doi​.org/​10​.1371/​ journal​.pone​.0039333. Timmerman, H., L. Mulder, H. Everts, D. Van Espen, E. Van Der Wal, G. Klaassen, S. Rouwers, R. Hartemink, F. Rombouts, and A. Beynen. 2005. Health and growth of veal calves fed milk replacers with or without probiotics. J. Dairy Sci. 88:2154–2165. https:​/​ /​doi​.org/​10​.3168/​jds​.S0022​-0302(05)72891​-5. Urie, N., J. Lombard, C. Shivley, C. Kopral, A. Adams, T. Earleywine, J. Olson, and F. Garry. 2018. Preweaned heifer management on US dairy operations: Part V. Factors associated with morbidity and mortality in preweaned dairy heifer calves. J. Dairy Sci. https:​ /​/​doi​.org/​10​.3168/​jds​.2017​-14019. Urubschurov, V., P. Janczyk, R. Pieper, and W. B. Souffrant. 2008. Biological diversity of yeasts in the gastrointestinal tract of weaned piglets kept under different farm conditions. FEMS Yeast Res. 8:1349–1356. https:​/​/​doi​.org/​10​.1111/​j​.1567​-1364​.2008​.00444​.x. USDA. 2007. Heifer calf health and management practices on U.S. dairy operations. USDA Animal and Plant Health Inspection Service, Veterinary Services, Ft. Collins, CO. Van den Abbeele, P., T. Van de Wiele, W. Verstraete, and S. Possemiers. 2011. The host selects mucosal and luminal associations of coevolved gut microorganisms: A novel concept. FEMS Microbiol. Rev. 35:681–704. https:​/​/​doi​.org/​10​.1111/​j​.1574​-6976​.2011​.00270​ .x. Winder, C. B., D. F. Kelton, and T. F. Duffield. 2016. Mortality risk factors for calves entering a multi-location white veal farm in Ontario, Canada. J. Dairy Sci. 99:10174–10181. https:​/​/​doi​.org/​10​ .3168/​jds​.2016​-11345. Yu, Z., and M. Morrison. 2004. Improved extraction of PCR-quality community DNA from digesta and fecal samples. Biotechniques 36:808–812. https:​/​/​doi​.org/​10​.2144/​04365ST04.

Journal of Dairy Science Vol. 102 No. 8, 2019