EFFECT OF SACCHAROMYCES CEREVISIAE SUPPLEMENTATION ON HEALTH AND PERFORMANCE OF DAIRY COWS DURING TRANSITION AND EARLY LACTATION PERIOD

Data concerning the effect of probiotics supplementation on many parameters concurrently at the same cows are lacking. Therefore, the objective of this experiment was to investigate the effects of Saccharomyces cerevisiae feeding on rumen, blood and milk parameters together in highproducing dairy cattle during the transition and early lactation period. Sixteen clinically healthy Holstein cows were divided into 2 groups: a control group of 6 cows and a probiotics-fed group of 10 cows. Rumen fluid and blood samples were collected 21 days before the expected calving as well as 7, 15, 30, 45 and 60 daysin-milk (DIM). Milk yield for each animal was recorded every 2 weeks. Individual milk samples were collected 15, 30, 45 and 60 DIM. Ruminal pH and rumen ammonia nitrogen were significantly lower, whereas total volatile fatty acids were significantly higher in yeast-fed animals compared with controls throughout the study. Serum concentrations of total proteins and globulins were higher, while albumins were lower in the yeast-treated group. Serum glucose levels were significantly higher in yeast-supplemented animals. Serum triglycerides, high density lipoproteins, and low density lipoproteins concentrations were lower, with cholesterol being significantly lower in the treated group. Milk production and milk fat percentage were higher, whereas milk protein percentage and somatic cell count were decreased in yeast-supplemented cows throughout the study. These results suggest that supplementation of S. cerevisiae to dairy cows rations during transition and early lactation period improve their health and milk production parameters.


Introduction
The transition period of a dairy cow is defined as the change from the pregnant, non-lactating state to the non-pregnant, lactating state; it lasts from 3 weeks pre-partum until 3 weeks postpartum (Goff and Horst, 1997).It is characterised by numerous changes in physiological, metabolic and endocrine status to accommodate parturition and lactogenesis (Grummer, 1995).If nutritional management does not meet these challenges, the transition cow is at high risk of developing a wide range of health problems soon before and, mainly, after parturition (Bell, 1995), like milk fever, fatty liver, ketosis, retained placenta, displaced abomasum, and suppressed immune function (Goff and Horst, 1997).
Probiotics are beneficial for animals, affecting their health and production by various mechanisms that are not yet fully understood (Shriver-Munsch, 2011).Feeding yeast (Saccharomyces cerevisiae) or its fermentation products during the transition period may counteract some of those challenges by improving appetite, nutrient utilisation and immune function (Shriver-Munsch, 2011).Yeast culture used as a dietary supplement for dairy cattle is thought to improve rumen function, and hence milk production and feed efficiency, by stimulating selective growth of rumen bacteria species (Harrison et al., 1988).
Inclusion of S. cerevisiae in ruminants' diets has been shown to alter the molar proportion of ruminal volatile fatty acids (VFAs) (Newblod et al., 1990;Dawson, 1993), reduce rumen ammonia concentration, increase the number of ruminal bacteria and protozoa and alter the flow of the nitrogen (N) fraction to the duodenum (Dawson, 1993;Williams et al., 1991).Furthermore, a study by Kumar et al. (1994) showed that supplementation of yeast culture as a growth promoter for buffalo calves resulted in increased rumen pH, total bacteria and protozoa culture counts, total VFAs, total N and microbial protein, with reduced rumen ammonia N concentration and improved digestion of cellulose and dry matter (DM) intake.Other researchers have reported that live yeast and yeast culture supplementation may increase feed intake and milk production of dairy cows (Robinson and Garrett, 1999;Dann et al., 2000).
However, each published study has investigated the effect of S. cerevisiae feeding to dairy cows selectively on a few parameters (either milk yield or milk composition, or either blood biochemical or ruminal parameters) and, therefore, data concerning its effect on many of the parameters together are lacking.The objective of the present study was to investigate the effect of feeding S. cerevisiae concurrently on rumen parameters, blood biochemical parameters, milk production, and milk composition in high-producing dairy cows during the transition and the early lactation period.

Experimental design, feeding and management
The present study was carried out in a private farm in Ihnasia city, Beni-Suef governorate in Egypt.Sixteen clinically healthy Holstein cows aged 4-5 years old, with an average body weight of 530±22 kg (mean±SE) were used.The animals were randomly allocated into 2 groups that were similar according to parity, body weight and previous mean total milk yield.The first (group A) was consisted of 6 animals, fed on a diet without yeast supplementation and kept as a control group.The second (group B) was consisted of 10 cows, fed on the same diet as group A plus daily in-feed inclusion of 0.5 g/animal of live yeast culture (Levucell SC 20 ® ).The animals were housed in a clean and spacious open yard.Cows were milked by an automated milking machine three times daily and fed on a total mixed ration to meet the recommendations of the National Research Council (NRC, 2001).They were accustomed to 2 diets, the close-up dry period diet (fed 3 weeks before expected calving up to the day of parturition) and the early lactation diet (fed from parturition up to 60 days-in-milk -DIM).The ingredients and nutrient composition of the 2 diets are shown in Tables 1 and 2.

Samplings and analyses
Rumen and blood samples were collected at the commencement of the experiment (21 days before the expected calving date), as well as 7, 15, 30, 45 and 60 DIM.All animals were clinically examined before each sampling.

Rumen fluid
Rumen fluid was collected from all animals using stomach tubing, 4 hours after the morning feeding.At each sampling 100 mL of rumen fluid were collected into a clean, dry flask.Ruminal pH was immediately measured using a portable digital pH meter (350 portable pH meter, JENWAY, Essex, UK).Rumen fluid samples were kept frozen (-20˚C) for analysis.Rumen ammonia nitrogen concentration was determined according to the method proposed by Conway (1974) and total VFAs samples were assayed by steam distillation, according to the method described by Abou-Akkada and El-Shazly (1964).

Blood samples
Blood samples were collected by jugular venipuncture, 2 hours after the morning feeding.Samples were centrifuged at 4000 rpm for 15 min to obtain blood sera, which were stored at -20ºC until analysis.
Serum was tested for total proteins (TP) and albumins (ALB) according to Doumos et al. (1971), and then serum globulins (GLOB) were calculated.Serum glucose levels were estimated according to Trinder (1969).Concentrations of serum total cholesterol, triglycerides, high density lipoproteins (HDL) and low density lipoproteins (LDL) were measured using commercial kits (SPECTRUM DIAGNOSTICS, Obour City, Egypt).

Milk samples
Milk yield of each animal was recorded every 2 weeks.Individual milk samples of each animal were collected for analysis at 15, 30, 45 and 60 DIM.Fat percentage, protein percentage and somatic cell count (SCC) were measured at Animal Reproduction Research Institute using MilkoScan analyser (FOSS ANA MilkoScan FT 120, GERBER INSTRUMENTS, Effretikon, Switzerland), according to the method proposed by Zecconi et al. (2002).

Statistical analysis
Data were statistically analysed using SAS computer software (SAS, 1985).The general linear model function was used for analysis of variance (ANOVA).Statistically significant differences between treatment means were measured by least significant difference and means were considered different at (P < 0.05) and at (P < 0.01).

Results
All 16 animals remained clinically healthy during the whole experimental period.There was not any statistically significant difference between any of the investigated parameters at the commencement of the experiment (21 days before the expected calving) between experimental groups.
Rumen fluid parameters for control and S. cerevisiae-fed animals are shown in Table 3. Ruminal pH was significantly lower at 15 (P<0.05),45 (P<0.01) and 60 (P<0.05)DIM in yeast-fed animals compared to control ones.The levels of rumen ammonia nitrogen in yeast-treated cows were significantly lower than controls at 15 (P<0.05),30 (P<0.01) and 45 (P<0.01)DIM.Total VFAs concentrations were significantly higher in treated animals compared to the control cows throughout the experiment.As shown in Table 4, serum TP concentrations tended (P>0.05) to be higher in the yeast-supplemented group B than those in the control group A, except for the 45 th DIM where they were lower.On the contrary, serum ALB values showed a non-significant (P>0.05)reduction in the yeast-supplemented group compared with control ones throughout the study.GLOB concentrations tended (P>0.05) to be higher in the yeast-supplemented group, except for the 60 th DIM where a significant difference (P<0.01) was noted.Serum glucose levels were significantly higher (P<0.01) in yeastsupplemented animals than controls at 30, 45 and 60 DIM.Concerning the effects of probiotics upon serum lipids, a significant reduction (P<0.05) in the levels of triglycerides was evident at 60 DIM in S. cerevisiae-supplemented group.HDL levels were significantly lower (P<0.05) in yeast-fed group at 45 DIM, while for LDL no significant differences were observed between 2 groups during the whole study period.Furthermore, a significant reduction was found for serum cholesterol concentrations in the yeast-fed group compared to the control group at 30 (P<0.01), 45 (P<0.01) and 60 (P<0.05)DIM.
Regarding milk production and milk parameters (Table 5), the results revealed that milk yield was higher in yeast-supplemented group, but it was significantly higher (P<0.01)only at 60 DIM.Milk fat percentage was higher (P>0.05) in yeast-fed group throughout the study.In contrast, a significant reduction (P<0.01) was evident for milk protein percentage at 45 and 60 DIM in yeast-fed animals.Similarly, SCC was lower in this group, but significantly lower (P<0.01)at 15 DIM.

Discussion
The results showed that rumen pH was lower in yeast-fed animals compared to controls.These results are in accordance with those found by other workers (Herrick,1971;Khattab et al., 2003).The reduction in the rumen pH of yeast-fed animals may be attributed to the increase in the concentration of VFAs production (Hristov et al., 2001;Ghorbanietal., 2002) and/or the increase in lactate synthesis in probiotics-supplemented animals (Benjamin 1990).The fact that the total VFAs concentration in our study were higher in the yeast-fed group throughout the experimental period is supportive for the hypothesis of Hristov et al. (2001) andGhorbani et al. (2002).The present results concerning total VFAs are consistent with those obtained by previous studies (Windschtil, 1991;1998;Agarwal et al., 2002;Abd El-Tawab, 2007).The increase of total VFAs concentrations in yeast-supplemented animals may be attributed to decreased methane production and consequent reduction of energy loss, providing thus additional energy for VFAs synthesis (Williams and Newbold, 1990).
Concentrations of rumen ammonia nitrogen were significantly lower in the yeast-treated animals.This is in agreement with previously published results (Harrison et al., 1988;Carro et al., 1992;Erasmus et al., 1992).The reduction of rumen ammonia nitrogen in yeast-supplemented animals may be attributed to the incorporation of ammonia into microbial proteins rather than a decrease in protein degradation (Williams and Newbold, 1990), or it may be due to an inhibitory effect on proteolysis, amino acid production, or ruminal urease activity (Khattab et al., 2003).
Concentrations of serum TP and GLOB were non-significantly higher in yeast-supplemented group.Similar results were obtained by other researchers (Abd El-Tawab, 2007;Helal and Abdel-Rahman, 2010).The elevation in the levels of serum TP in probiotics-treated animals may be attributed to the fact that yeast supplementation stimulates the rumen microbial protein synthesis, so it elevates the populations and the activity of cellulolytic bacteria in rumen, consequently enhancing the fiber digestion, lactate utilization in the rumen and increase flow of microbial protein from the rumen to duodenum (Guedes et al., 2008).Moreover, the elevated levels of GLOB in the S. cerevisiae-fed group may be due to the increase of net globulins, as a result of the increase in gamma globulins caused by Kupffer cell proliferation and an increase in the number of plasma cells in the bone marrow (Benjamin, 1984).This hypothesis is supported by other researchers (Buts et al., 1990), who found that oral administration of S. cerevisiae to growing rats significantly increased IgA and secretory components of immunoglobulins.
Serum ALB values were not significantly different between the 2 groups, although a slight reduction was observed in the yeast-supplemented group.This reduction might be attributed to the increased milk production of these animals (Benjamin, 1984;Helal and Abdel-Rahman, 2010).Results of milk production in the present study favour this hypothesis.
The finding that serum glucose levels were significantly higher in yeastsupplemented animals is in agreement with the results of other studies (Abd El-Tawab, 2007;Kawas et al., 2007;Stanislaw and Przemyslaw, 2009).The glucose increase in yeast-treated animals may be attributed to increased gluconeogenesis, which raises blood glucose levels in ruminants (Huntington and Eisemann, 1988).This explanation is supported by Antunovic et al. (2005),who recorded low glucose levels in probiotics-treated lambs after inhibition of gluconeogenesis by insulin, which inhibits phosphorylase and gluconeogenic enzymes.It is worth mentioning that propionate is the major precursor for gluconeogenesis in ruminants and thus increments of rumen propionate production result in an increase of hepatic glucose production (Reynolds et al., 2003;Stein et al., 2006).The higher propionic acid production is the cause of the increased glucose levels in S. cerevisiae-fed cows (Nisbet and Martin, 1991).Several studies demonstrated that feeding S. cerevisiae increased the production of acetate, propionate, and total VFA in dairy cows (Nisbet and Martin, 1991;Piva et al., 1993;Miller-Webster et al., 2002).Total VFAs were significantly higher for yeast-fed cows in our experiment and this could explain the recorded higher glucose levels in yeast-treated cows.
Serum triglycerides, HDL, LDL and cholesterol were lower in the yeastfed group compared with the control one.This reduction may be due to an increase in lipid metabolism and utilisation by the cows because of their increased milk production (Stein et al., 2006), which was the case for the group of the yeastsupplemented cows in our study.Moreover, the reduction in cholesterol level could be attributed to the inhibition of the cholesterol synthesis or the direct assimilation of cholesterols (Zacconi et al., 1992).The lipid profile estimated in the current study is in harmony with that reported by other researchers (Taranto et al., 1998;Begely et al., 2006).
The higher milk yield recorded for the yeast-supplemented cows in our experiment has also been observed by other workers and may be attributed to an increase in DM intake (Robinson andGarrett, 1999, Jounay, 2006), a greater flow of microbial protein and amino acids to the duodenum (Erasmus et al., 1992), and the fact that yeast supplementation may act as a source of vitamin B complex (Abdel-Khalek, 2003;Helal and Abdel-Rahman, 2010).However, some other researchers (Erdman and Sharma, 1989;Arambel and Kent, 1990;Kung et al., 1997) have not found probiotic administration to increase the milk production in cows.
The results for milk fat percentage in the present study are in agreement with those recorded by others (Oetzel et al., 2007;Hanafy et al., 2009;Metha et al., 2011).The increase in milk fat percentage in yeast-supplemented animals may be attributed to the increment in total bacterial populations and cellulolytic microorganisms in the rumen, which improve fiber digestibility and fermentation and, consequently, increase milk fat content (Doreau and Jounay, 1998;Wang, 2001;Chaucheyras-Daurant et al., 2008).
Milk protein content increases after probiotic administration according to some researchers (Abdel-Khalek, 2003;Helal and Abdel-Rahman, 2010) and decreases according to others (Lehloenya, et al., 2008).In the present experiment, a reduction of milk protein percentage was recorded after the first 15 DIM for yeast-fed cows.This reduction could be interpreted on the base of dilution effect of the higher milk production in yeast-fed cows (Lehloenya et al., 2008).On the other side, milk protein percentage was higher in yeast-treated cows at 15 DIM.These findings show that the effect of probiotics on milk protein concentration needs further investigation.
SCC in yeast-fed cows was lower compared with that of controls.These results are in agreement with other studies (Stein et al., 2006;Sretenović et al., 2008).The reduction of SCC in yeast-treated cows may be attributed to a better health status of their udder (Sretenović et al., 2008) or may be due to an improvement of the immune status of the yeast-supplemented cows, as a result of the increase in IgA and secretory components of immunoglobulins (Buts et al., 1990).The increased values of globulins recorded in the yeast-treated group in our experiment could support this hypothesis.

Conclusion
Feeding S. cerevisiae to dairy cows during the close-up dry and the early lactation period significantly reduced rumen pH and rumen ammonia nitrogen, while significantly increased rumen total VFAs content.Serum glucose was significantly increased in the yeast-supplemented animals.Serum concentrations of total proteins and globulins were higher, while albumins were lower in the yeasttreated group.Serum lipids were significantly reduced in the yeast-fed group, except for LDL.Finally, milk production and milk fat percentage were higher, whereas milk protein percentage and SCC were lower in the S. cerevisiaesupplemented cows throughout the experiment.

CONFLICT OF INTEREST STATEMENT
The authors have nothing to disclose.

Table 3 . Effect of feeding S. cerevisiae during transition and early lactation period on rumen fluid parameters of Holstein dairy cows.
* Significantly different (P<0.01);** Significantly different (P<0.05)

Table 5 . Effect of feeding S. cerevisiae during the transition and early lactation periods on milk production and milk parameters of Holstein dairy cows.
* Significantly different (P<0.01);** Significantly different (P<0.05)