Oral administration of lactate producing bacteria alone or combined with Saccharomyces cerevisiae and Megasphaera elsdenii on performance of fattening lambs

ABSTRACT

To investigate the effect of direct fed microbial (DFM) on performance of fattening lamb, 4 treatments (7 animals/treatment) were evaluated; (1) control (without additive; CON); (2) Lactobacillus fermentum and Lactobacillus plantarum (FP); (3) Saccharomyces cerevisiae (SC) plus FP (SCFP) and (4) Megasphaera elsdenii plus SCFP (MSCFP). Feed intake, final body and feed efficiency except average daily gain were not affected by experimental treatment. Digestibility of dry matter and organic matter increased (P < 0.05) in MSCFP lambs versus other treatments. The highest (P < 0.05) amount of N intake, absorbed N and N retention were observed in MSCFP lambs. Blood parameters and enzymatic profiles were not affected (P > 0.05) by the treatments. The lowest and the highest of minimum pH were observed in CON and SCFP lambs. Moreover, concentration of ammonia-N, total and individual volatile fatty acids (except acetate and butyrate concentration) and acetate to propionate ratio were not affected (P > 0.05) by treatment. According to the results, the use of microbial feed additives had a positive effect on growth performance and some digestive and ruminal fermentation parameters.

KEYWORDS:

• Blood metabolite
• digestibility
• growth performance
• lamb
• microbial additives
• ruminal fermentation

1. Introduction

Increased concern about the use of antibiotics has increased interest in the use of natural feed additives such as direct-fed microbial (DFM). The DFM are viable organisms that have been shown positive effects on animal health and performance when used as dietary additive (Arowolo and He 2018; Rodríguez-Gaxiola et al. 2020). Preliminary research into the use of microbial inoculants was conducted by Allison et al. (1964). They used the rumen content of high-grain-adapted animal as microbial inoculants to prevent ruminal acidosis. Several microorganisms known as non-pathogenic are used as DFM such as Lactobacillus spp., Bifidobacterium, Enterococcus, Bacillus spp., lactic acid utilizing bacteria (LUB) and different strains of yeast (Seo et al. 2010). The purpose of using DFM is to improve productive performance and prevent disease by maintaining gastrointestinal health and improving gut function (Chaucheyras-Durand et al. 2008). In addition, DFM enhance rumen microbial ecosystem (Musa et al. 2009; Kewan et al. 2019), reduce greenhouse gases emission (Hernandez et al. 2017; Vallejo-Hernández et al. 2018; Pedraza-Hernández et al. 2019) increase digestibility (El-Ghani 2004; Kewan et al. 2019), nutrient absorption and improve feed conversion ratio (FCR) (Antunović et al. 2006; Whitley et al. 2009), animal performance and prevent the accumulation of lactic acid (Jouany and Morgavi 2007).

Research has shown that the use of lactic acid-producing bacteria (LAB) as bacterial DFM has a positive effect on LUB and prevent ruminal acidosis (Goto et al. 2016). Bacterial DFM containing LAB have been reported to stabilize ruminal microbial flora (Chiquette et al. 2012), which in turn leads to increased feed intake, weight gain and improved animal health (Timmerman et al. 2005). Among the LAB, Lactobacillus plantarum and Enterococcus faecium are mainly used as bacterial DFM (Weinberg et al. 2003). The LUB such as Megasphaera elsdenii, Propionibacterium and Selenomonas ruminantium can play a key role in controlling ruminal fermentation (Mackie and Gilchrist 1979). M. elsdenii and S. ruminantium are the predominant strains of LUB in the rumen (Henning et al. 2010). It has been estimated that M. elsdenii utilize 65% to 95% of the lactic acid in the rumen and leads to prevents a sharp decrease in ruminal pH as a result of the accumulation of lactic acid (Kung and Hession 1995). Thus, the LUB can change lactate to volatile fatty acids (VFA), increase the production of propionate compared to lactic acid, increase feed efficiency and increase ruminal pH (Seo et al. 2010). Yeast additives lead to an increase the population of rumen bacteria, especially cellulolytic bacteria, through the equilibrium in rumen pH (Beauchemin et al. 2006). Hence, live yeast and yeast culture lead to increase the population of rumen bacteria, especially cellulolytic bacteria, through the equilibrium in rumen pH (Beauchemin et al. 2006). Therefore, improving ruminal digestion as a result of feeding with yeast additives is associated with increased pH as well as VFA (Dolezal et al. 2005).

The impact of using DFM on performance and ruminal fermentation in ruminant has been investigated in several studies (Vallejo-Hernández et al. 2018; Kewan et al. 2019; Pedraza-Hernández et al. 2019), and according to Nocek and Kautz (2006), the use of DFM in dairy cattle has become a commonplace occurrence. The most used commercial products in feeding dairy cows include LAB strains (Jin et al. 2000; McAllister et al. 2011) alone or in combination with Saccharomyces cerevisiae (SC). Although the use of yeast in feeding small ruminants has been investigated in several articles (Abdelrahman and Hunaiti 2008; Titi et al. 2008; Soren et al. 2013), few studies have been conducted on the use of bacterial DFM in sheep. However, this study was performed to investigate the effect of LAB (Lactobacillus fermentum, Lactobacillus plantarum (FP)) alone or combined with LUB (M. elsdenii (Me)) and SC on growth performance, nitrogen (N) balance, in vivo nutrient digestibility, blood metabolites and rumen fermentation parameters in fattening lambs.

2. Materials and methods

2.1. Animals, housing, diet and treatments

All animal management and sampling procedures were conducted according to The Care and Use of Agricultural Animals in Research and Teaching guidelines (FASS 2010). All procedures and guidelines involving animals were approved by the Animal Experiment Committee at Agricultural Sciences and Natural Resources University of Khuzestan, Mollasani, Iran.

Twenty-eight autumn born Arabian lambs (male) with average weight 24 ± 3.7 kg and age 120 days old were used in a completely randomized design with 4 treatments (DFM) and 7 replicates. The lambs were randomly divided into 4 groups and housed in individual pens (1.5 m × 1.3 m). The experiment duration was 77 days (14 days for adaptation and 63 days was the main period of the experiment). During the adaptation period, all lambs were vaccinated against external (Azantole) and internal (Triclabendazole and Levamisole) parasites. Lambs received a diet containing 70% concentrate and 30% forage that formulated according to NRC (2007) – (Table 1). The animals were fed with the diet (total mixed ration) ad libitum, twice daily (at 08:00 and 16:00) to ensure remains 5% residual during 24 h. The lambs also had free access to water and salt licks during the experiment.

Table 1. Ingredients (g/kg DM), chemical composition (g/kg DM) and metabolizable energy (ME; Mcal/kg DM) of the experimental diet.

Ingredients
Alfalfa	201
Wheat straw	99
Barley grain	300
Corn grain	210
Soybean meal	123.5
wheat bran	55
Calcium carbonate	4
NaCl	2.5
Vitamin and mineral premix^a	5
Chemical composition^b
Dry matter	903
Organic matter	948
Crude protein	161
Ether extract	27
NDFom	290
ADFom	165
Lignin	30
ME^c	2.65
Non-fibre carbohydrates^d	472

Note: NDFom, ash-free neutral detergent fibre; ADFom, ash-free acid detergent fibre.

^aPremix contained (per kg): Vitamin A, 500,000 IU/mg; vitamin D₃, 100,000 IU/mg; vitamin E, 100 mg/kg; Ca, 180 g/kg; P, 60,000 mg/kg; Na, 60,000 mg/kg; Mg, 19,000 mg/kg; Zn, 3000 mg/kg; Fe, 3000 mg/kg; Mn, 19,000 mg/kg; Cu, 300 mg/kg; Co, 100 mg/kg; Se, 1 mg/kg; I, 100 mg/kg; antioxidant, 400 mg/kg; carrier, up to 1000 g.

^bAnalysed from each feed composition (explained in Section 2.2.).

^cEstimated from each feed composition.

^dCalculated as: 1000 – (NDFom g/kg + crude protein g/kg + ether extract g/kg + ash g/kg).

In treatment containing microbial additives, the lamb was inoculated with a 50 mL microbial suspension before morning feeding (daily oral dosed). In the control, 50 mL of water was dosed instead of the microbial suspension. Three microbial additives FP, SC and Me were used in this study and 4 treatments were evaluated according to the type of additive used: (1) control (without additive (only diet); CON), (2) FP (50 mL bacterial suspension containing 4.5 × 10⁸ cfu/day of L. plantarum and L. fermentum (in ratio 50:50)), (3) SC + FP (50 mL microbial suspension containing 4.5 × 10⁸ cfu/day FP and 1.4 × 10¹⁰ cfu/day SC; SCFP), and (4) Me + SCFP (50 mL microbial suspension containing 4.5 × 10⁸ cfu/day Me, 4.5 × 10⁸ cfu/day FP and 1.4 × 10¹⁰ cfu/day SC; MSCFP). The commercial strain of L. plantarum and L. fermentum (GP10) isolated from the rumen of Najdi goat were used as LAB. M. elsdenii (GU1) isolated from the rumen of Najdi goat was used as LUB. The yeast additive was used from Iran Mollasses company (Mashhad, Iran), that each gram of this yeast contains 7 × 10⁹ cfu/g.

2.2. Feed intake, growth performance, nutrient digestibility, N retention and chemical analysis

During the experiment period for estimating the voluntary feed intake, feed offered and residue were recorded daily before morning feeding, and feed and residue samples were stored at −20°C for subsequent chemical analysis. In order to determine the average daily gain (ADG), and feed efficiency (FE) (gain:feed), the lambs were weighed once every 14 days before morning feeding. The average gain on a daily basis was calculated for each lamb by linear regression analysis of body weight vs. time.

Digestibility coefficients of dry matter (DM), organic matter (OM), crude protein (CP), ash-free neutral detergent fibre (NDF) and ash-free acid detergent fibre (ADF) were estimated using the total faecal collection method (Givens et al. 2000). For this purpose, on day 50 of the experiment, 5 lambs from each treatment with the same body weight were transferred to metabolism crate equipped with faeces and urine collectors. The digestibility test lasted 13 days, with 7 days for adaptation and 6 days for sampling. During the collection period, the amount of feed offered, residue, faeces and urine of each lamb was recorded within 24 h. Each day 10% of samples (feed, residue, faeces, and urine) were collected and frozen at −20°C and at the end of the collection period daily samples were pooled of each lamb and thoroughly mixed and then frozen at −20°C for subsequent analysis.

Before the chemical analysis, feed, residue and faeces samples were oven-dried at 55°C for 48 h and then were ground using a mill equipped with a 1-mm sieve (Wiley mill, Swedesboro, USA). After that following the AOAC International (1998) procedure samples were analysed for CP (Number. 988.05), ether extract (EE) (Number. 920.39), ash (Number. 924.05) and ADF (Number. 973.18). Also, NDF was analysed according to Van Soest et al. (1991). Non-fibre carbohydrates (NFC) concentration was calculated by difference as NFC (g/kg DM) = 1000 − (NDF g/kg DM + CP g/kg DM + EE g/kg DM + ash g/kg DM).

Urinary N was also estimated using the Kjeldahl method. Daily N retention was calculated from the difference between the amount of daily N intake and the amount of daily N excreted (total N excreted in urine and faeces).

2.3. Blood sampling and analysis

In order to investigate blood biochemical parameter, blood sample (approximately 10 mL) was collected from jugular veins using tubes (Becton Dickinson, Rutherford, NJ, USA) containing an anticoagulant (heparin) on days 1 (the first day after finishing the adaptation), 14, 28, 42, 56 and 63 of the experimental period just 4 h after morning feeding. The blood samples were centrifuged (3,000 × g for 15 min at 4°C) and the plasma was separated and frozen at −20°C until measuring biochemical parameters. Glucose, triglycerides, total protein, albumin, creatinine, blood urea nitrogen (BUN), cholesterol, low-density lipoprotein (LDL), high-density lipoproteins (HDL), serum glutamic pyruvic transaminase (SGPT) and serum glutamic oxaloacetic transaminase (SGOT) were determined by using enzymatic methods and spectrophotometer (model S Bio-Rad Libra, England) and using kits of the ZiestChem Company (Tehran, Iran).

2.4. Ruminal fermentation parameter

In order to investigate ruminal pH changes during the experimental period on days 1 (the first day after finishing the adaptation), 14, 28, 42, 56 and 63 and 0, 3 and 6 h after morning feeding the rumen fluid was obtained with a stomach tube. The Min and Max of pH were determined by mean of Min and Max pH during sampling days. Ruminal pH was determined immediately by a portable pH meter (Metrohm model, Swiss). To investigate ruminal VFA and ammonia-N, the ruminal fluid that obtained on day 63 (0, 3 and 6 h after morning feeding) was strained through two layers of cheesecloth, and rumen fluid (25 mL) immediately submitted for the evaluation of ammonia-N with 5 mL of HCl 0.2 N and stored at −20°C. The rumen ammonia-N concentration was measured by phenol-hypochlorite assay (Broderick and Kang 1980). For analysis of ruminal VFA, 2 mL of strained rumen fluid was preserved, at −20°C, with 0.5 mL of an acid solution containing 200 mL/L orthophosphoric acid and 2-ethyl-butyric acid as internal standard. After thawing, rumen fluid samples were centrifuged (14,000 × rpm for 15 min; 4°C) and the supernatant was used to quantify the VFA. Ruminal VFA were determined by gas chromatography (GC; Chrompack, Model CP-9002, Chrompack, EA Middelburg, Netherlands) that equipped with a 50-m (0.32 mm ID) silica-fused column (CP-Wax Chrompack Capillary Column, Varian, Palo Alto, CA, USA). The helium was used as a carrier and oven initial and final temperatures were 55°C and 195°C, respectively, and detector and injector temperatures were set at 250°C.

2.6. Data analysis

The data obtained from assessing growth performance (7 lambs for each treatment), N balance (5 lambs for each treatment), nutrient digestibility (5 lambs for each treatment) and ruminal fermentation parameter (7 lambs for each treatment) were analysed as a randomized complete design using General Linear Models (GLM) procedure in SAS software (SAS 2008) based on the following statistical model:$Y_{ij}=\mu +T_i+e_{ij}$where Y_ij is observation, µ is the general mean, T_i is the effect of microbial additives and e_ij is the standard error of term. Also, data obtained from ruminal pH (5 lambs for each treatment) and blood metabolites (5 lambs for each treatment) were analysed as repeated measurements using the MIXED procedures of SAS based on the statistical model:$Y_{ijk}=\mu +T_i+H_j+(TH{)}_{ij}+e_{ijk}$

where Y_ijk is observation, µ is the general mean, T_i is the effect of microbial additives, H_j is effect of sampling day, (TH)_ij is interactions between effect of microbial additives and sampling day and e_ijk is the standard error of term. Means were compared by the Duncan multiple comparison tests at P < 0.05.

3. Results

3.1. Feed intake and growth performance

Intake of DM, OM, CP, NDF and ADF were not affected (P > 0.05) by dietary supplementation with different microbial additives (Table 2). In addition, the final body weight and FE were not affected (P > 0.05) by dosed DFM. However, the ADG increased (P < 0.05) in MSCFP versus CON lambs.

Table 2. Feed intake and growth performance of lambs fed with different sources of microbial additives.

Item	Treatments				SEM	P-value
	CON (n = 7)	FP (n = 7)	SCFP (n = 7)	MSCFP (n = 7)
Intake (g/d)
Dry matter	1021	1029	1054	1070	30.7	0.69
Organic matter	970	977	1001	1016	28.8	0.68
Crude protein	165	166	170	173	4.70	0.67
NDFom	294	296	302	308	9.79	0.77
ADFom	167	170	172	176	5.74	0.76
Performance
Initial body weight (kg)	24.3	23.8	24.5	24.4	1.62	0.95
Final body weight (kg)	37.3	37.3	38.2	38.3	1.64	0.95
Average daily gain (g/d)	205^b	214^ab	218^ab	220^a	6.82	0.047
Feed efficiency (Gain:Feed)	0.202	0.209	0.208	0.207	0.01	0.94

Notes: NDFom, ash-free neutral detergent fibre; ADFom, ash-free acid detergent fibre; CON, without microbial additive; FP, Lactobacillus fermentum and Lactobacillus plantarum; SCFP, Saccharomyces cerevisiae (SC) plus FP; MSCFP, Megasphaera elsdenii plus SC plus FP; SEM, standard error of means. ^a-b Means in the same row with different superscript letters are different (P < 0.05).

3.2. Nutrient digestibility, N balance and blood metabolites

Digestibility of DM and OM increased (P < 0.05) in MSCFP lambs versus other treatments. The highest (P < 0.05) digestibility of NDF was observed in the MSCFP lambs, but there was no difference with SCFP and CON lambs. However, there were no effect (P > 0.05) among treatments on the digestibility of CP, ADF and ME (Table 3).

Table 3. Apparent digestibility (%) and estimated metabolizable energy (Mcal/kg DM) of lambs fed with different sources of microbial additives.

Apparent digestibility	Treatments				SEM	P-value
	CON (n = 5)	FP (n = 5)	SCFP (n = 5)	MSCFP (n = 5)
Dry matter	73.2^b	72.1^b	73.3^b	75.6^a	1.11	0.041
Organic matter	75.6^b	74.9^b	75.8^b	78.6^a	0.87	0.013
Crude protein	78.5	77.2	78.9	78.8	0.76	0.361
NDFom	53.7^ab	51.2^b	53.4^ab	55.4^a	1.50	0.017
ADFom	51.4	50.7	53.8	56.1	2.67	0.436
ME	2.69	2.67	2.70	2.75	0.04	0.317

Notes: NDFom, ash-free neutral detergent fibre; ADFom, ash-free acid detergent fibre; ME, metabolizable energy; CON, without microbial additive; FP, Lactobacillus fermentum and Lactobacillus plantarum; SCFP, Saccharomyces cerevisiae (SC) plus FP; MSCFP, Megasphaera elsdenii plus SC plus FP; SEM, standard error of means. ^a,b Means in the same row with different superscript letters are different (P < 0.05).

The highest (P < 0.05) DM and OM intake and also digestible organic matter intake (DOMI) were observed in MSCFP lambs during the sampling week (days 57–63). Moreover, for MSCFP lambs, N intake was not higher than SCFP treatment and N retention was not higher than FP and SCFP lambs (Table 4).

Table 4. N balance of lambs fed with different sources of microbial additives.

Item	Treatments				SEM	P-value
	CON (n = 5)	FP (n = 5)	SCFP (n = 5)	MSCFP (n = 5)
Intake (g/d)
Dry matter	1017^c	1087^b	1122^a	1158^a	17.9	0.006
Organic matter	946^c	1013^b	1044^a	1077^a	16.5	0.006
DOMI	720^c	758^bc	791^ab	833^a	19.6	0.010
N balance
N Intake (g/d)	26.2^b	27.1^b	28.9^a	29.8^a	0.43	0.005
Faecal N (g/d)	5.54^b	6.45^a	6.15^a	6.33^a	0.16	0.003
Urinary N (g/d)	6.32^c	5.13^d	7.89^a	7.26^b	0.19	0.001
N retention
Total N retention (g/d)	14.4^c	15.5^a	15.0^a	16.2^a	0.51	0.006
mg/g of N intake	540^b	573^a	517^c	544^b	9.19	0.013
mg/g of N digested	686^b	752^a	655^c	690^b	7.28	<0.0001

Notes: Intake, feed intake during of sampling; DOMI, digestible organic matter intake; N, nitrogen; CON, without microbial additive; FP, Lactobacillus fermentum and Lactobacillus plantarum; SCFP, Saccharomyces cerevisiae (SC) plus FP; MSCFP, Megasphaera elsdenii plus SC plus FP; SEM, standard error of means. ^a,dMeans in the same row with different superscript letters are different (P < 0.05).

Plasma parameters and enzymatic profiles were not affected (P > 0.05) by the experimental treatments (Table 5).

Table 5. Blood chemistry parameters of lambs fed with different sources of microbial additives.

Item	Treatments				SEM	P-value
	CON (n = 5)	FP (n = 5)	SCFP (n = 5)	MSCFP (n = 5)		T	D	T × D
Glucose (mg/dL)	82.2	81.8	82.7	82.3	0.95	0.831	<0.0001	0.729
BUN (mg/dL)	18.5	19.2	21.5	21.3	1.16	0.190	0.0005	0.016
Creatinine (mg/dL)	0.75	0.81	0.81	0.80	0.03	0.747	0.0002	0.970
Triglycerides (mg/dL)	19.6	19.7	19.9	20.1	0.14	0.288	0.0008	0.801
Cholesterol (mg/dL)	58.4	56.4	58.1	55.8	1.70	0.925	0.132	0.566
LDL (mg/dL)	20.3	18.4	19.4	19.3	0.99	0.648	<0.0001	0.209
HDL (mg/dL)	25.1	23.8	22.7	24.4	1.83	0.181	<0.0001	0.0132
Total protein (g/dL)	7.13	6.76	6.68	6.36	0.30	0.221	0.294	0.407
Albumin (g/dL)	3.48	3.24	3.21	3.29	0.13	0.097	0.363	0.061
SGPT (U/L)	17.0	14.8	17.6	15.2	1.03	0.597	0.019	0.191
SGOT (U/L)	77.6	76.5	77.9	80.6	1.67	0.505	<0.0001	0.048

Notes: Average of repeated sampling of blood collected on days 1, 14, 28, 42, 56 and 63 of experimental period, just 4 h after morning feeding. BUN, blood urea nitrogen; LDL, low-density lipoprotein; HDL, high-density lipoproteins; SGPT, serum glutamic pyruvic transaminase; SGOT, serum glutamic oxaloacetic transaminase; CON, without microbial additive; FP, Lactobacillus fermentum and Lactobacillus plantarum; SCFP, Saccharomyces cerevisiae (SC) plus FP; MSCFP, Megasphaera elsdenii plus SC plus FP; SEM, standard error of means. T, effect of treatment; D, effect of sampling day.

3.3. Ruminal fermentation parameters

The lowest and the highest (P < 0.05) of minimum pH values were observed in CON and SCFP lambs, whereas the maximum pH was not affected (P = 0.443) by experimental treatments. There was no significant difference (P > 0.05) among the experimental treatments in the concentration of ammonia-N, total VFA, propionate, valerate, isobutyrate, isovalerate and acetate to propionate ratio. The highest (P < 0.05) concentration of acetate and butyrate was observed in CON and FP lambs, respectively, while the concentration of butyrate was not different (P > 0.05) between FP and SCFP (Figure 1 and Table 6.).

Figure 1. Effect of different source of microbial additives on ruminal pH of the lambs. □ (CON), without microbial additive (control); ▪ (FP), Lactobacillus fermentum and Lactobacillus plantarum; ▴ (SCFP), Saccharomyces cerevisiae (SC) plus FP; Δ (MSCFP), Megasphaera elsdenii plus SC plus FP. Arrow (↑), morning feeding at 8.

Table 6. Ruminal fermentation parameters of lambs fed with different sources of microbial additives.

Item	Treatments				SEM	P-value
	CON (n = 5)	FP (n = 5)	SCFP (n = 5)	MSCFP (n = 5)
pH
Min	5.82^c	5.93^b	6.08^a	5.99^b	0.034	0.009
Max	7.41	7.20	7.26	7.24	0.088	0.443
Mean	6.54	6.55	6.56	6.53	0.026	0.912
Ammonia-N (mg/dL)	22.1	26.0	25.4	24.8	1.77	0.479
Total VFA (mmol/L)	113	116	114	117	2.21	0.709
VFA as % of total
Acetate (A)	51.8^a	48.5^b	50.1^b	49.1^b	0.704	0.015
Propionate (P)	24.0	24.4	24.9	26.7	0.721	0.081
Butyrate	18.4^b	21.1^a	20.8^a	19.3^b	0.568	0.034
Valerate	2.69	2.91	2.88	2.70	0.116	0.428
Isobutyrate	1.31	1.30	1.35	1.34	0.032	0.781
Isovalerate	1.53	1.48	1.54	1.53	0.116	0.994
A:P	2.15	1.99	2.01	1.84	0.143	0.642

Notes: Data obtained from pH were analysed as repeated and this table we used only the main effect of treatment. Also, the volatile fatty acids and ammonia-N were analysed for each treatment on day 63, at 0, 3, and 6 h after morning feeding. Min and Max, average Min and Max on days 1, 14, 28, 42, 56 and 63 of experimental period. VFA, volatile fatty acid; CON, without microbial additive; FP, Lactobacillus fermentum and Lactobacillus plantarum; SCFP, Saccharomyces cerevisiae (SC) plus FP; MSCFP, Megasphaera elsdenii plus SC plus FP; SEM, standard error of means. ^a–cMeans in the same row with different superscript letters are different (P < 0.05).

4. Discussion

4.1. Feed intake and growth performance

Microbial additives lead to increases feed intake (Chiofalo et al. 2004; Antunović et al. 2006; Desnoyers et al. 2009) and it has been found that increased feed intake affects ruminant performance (Beigh et al. 2017). Hassan et al. (2019) reported that the use of microbial supplements (Ruminococcus flavefaciens) increased feed intake. Moreover, final weight, ADG, total gain and FCR were improved in lambs supplemented with Pediococcus acidilactici and Pediococcus pentosaceus after weaning compared to control (Saleem et al. 2017). In a meta-analysis have been shown (21 articles were reviewed between 1985 and 2010) by adding DFM containing LAB to milk replacer of calves lead to improved body weight gain and FE compared to controls (Frizzo et al. 2011). Overall, DFM by affecting on ruminal pH and nutrient digestibility may affect the amount of intake (Wallace and Newbold 1993). Improved animal performance due to the use of probiotics can also result in improved growth of fibre-degrading bacteria and increased digestion of fibre (Russell and Wilson 1996). Feed intake and growth performance, of the present study, were not affected by experimental treatments except the ADG. Higher weight gain in lambs dosed with microbial additives could be due to increased microbial protein synthesis and post-ruminal availability of amino acid (Erasmus et al. 1992). The lack of effect of microbial additives on performance was attributed to the high level of dietary protein (Titi et al. 2008). The addition of yeast to the low-protein diet had a favourable effect on the performance compared to the high-protein diet (Kawas et al. 2007). Similarly to our results, in several studies supplementation with LUB (Propionibacterium) and LAB (Enterococcus faecium) (Ghorbani et al. 2002) and yeast culture did not have any effect on feed intake (Haddad and Goussous 2005; Hernandez et al. 2009). Moreover, supplementing the diet with L. acidophilus (Abu-Tarboush et al. 1996; Cruywagen et al. 1996), a mixture of L. acidophilus and Streptococcus faecium (Higginbotham and Bath 1993), a mixture of L. acidophilus and L. plantarum (Abu-Tarboush et al. 1996), B. subtilis (Galina et al. 2009) or a mixture of L. acidophilus, L. lactis and B. subtilis (Galina et al. 2009) had no effect on calf growth performance.

4.2. Nutrient digestibility and N balance

Nutrient digestibility was not different among treatments CON, FP and SCFP, while the combination of all three additives increased the digestibility of DM, OM and NDF. The improved of nutrient digestibility may also be due to the release of the enzyme by DFM and changes in rumen microbial ecology. However, what is known is that the optimal function of DFM on digestibility involves altering the rate of acid production, creating favourable microbial flora, and increasing fibre digestion (McAllister et al. 2011). Soren et al. (2013) reported that lambs fed with a diet supplemented by combining S. cerevisiae with L. sporogenes had no effect on the digestibility of DM, OM, and NDF. In other experiments in dairy and beef cattle, it has also been reported that the use of LAB in combination with LUB did not affect nutrient digestibility (Ghorbani et al. 2002; Raeth-Knight et al. 2007). However, it has been reported that the combination of L. acidophilus NP51 and Propionibacterium freudenreichii NP24 in dairy cows improved the digestibility of CP, NDF and ADF (Boyd et al. 2011). In addition, use of Enterococcus faecium and yeast improved DM digestibility of forage in cows during the transition period (Nocek and Kautz 2006).

The increase in urinary N excretion in the SCFP and MSCFP lambs was probably due to higher N intake. Increased protein intake and intestinal absorption of more amino acids than needed for tissues and or ammonia-N absorbed from the rumen epithelium (rumen ammonia-N concentration was numerically higher in experimental treatments than CON) lead to increased urinary N excretion (Willms et al. 1991). Moreover, the concentration of BUN numerically was higher in lambs of SCFP and MSCFP. Kohn et al. (2005) stated that BUN concentration has a direct relationship with excreted N in the urine. However, the highest N retention was observed in MSCFP lambs that was proportional to the increase in ADG and total gain. That could reflect a greater intake of N and possibly reflect its use by tissue and protein deposition. It has been reported DFM alone or in combination lead to increase N intake (Mohamed et al. 2009) and ruminal ammonia-N (Jouany, Mathieu, Senaud, et al. 1998). However, several studies have reported that microbial additives had no effect on the amount of N intake, N retention, and N excretion through urine and faeces (Jouany, Mathieu, Bohatier, et al. 1998; Hernandez et al. 2009).

4.3. Blood metabolites

The concentration of total protein and albumin in this study was not affected by the experimental treatments, which was in agreement with previous studies (Galip 2006; Soren et al. 2013). The lack of effect of the experimental treatments on the plasma protein indicates an improvement in the nutritional status of the animals and the lack of use of the amino acids and deamination to obtain energy. Bruno et al. (2009) observed that microbial additives lead to better use of N by ruminal bacteria and, thus reduce the concentration of BUN. In present study, the concentration of BUN was not affected by the experimental treatments and numerically was higher in dosed lambs (as results of high protein intake). Since creatinine concentration is constant, an increase in its concentration indicates renal disease (Direkvandi and Kamyab Kalantari 2018). The concentration of creatinine was lower the normal range (1.2–1.9 mg/dL) of plasma concentration of creatinine of sheep (Radostitis et al. 2007).

As the propionate concentration was not affected by the experimental treatments, the concentration of plasma glucose was not different among treatments. This is may be due to increased propionate production through increased gluconeogenesis lead to increases plasma glucose concentration (Hussein 2014). It has been reported that DFM has a positive effect on energy balance as a result of improved metabolic status. Therefore, it results in a decrease in plasma lipids concentration (Baiomy 2011). However, in our study concentrations of triglyceride, cholesterol, LDL and HDL were not affected by the DFM. Similarly, to our results, the concentrations of SGPT and SGOT were not affected by DFM, which was agreement with previous studies (Alhidary et al. 2016; Mousa et al. 2019).

4.4. Ruminal fermentation parameters

Ruminal acidosis occurs when the ruminal pH is reduced to less than 5.6 due to the accumulation of lactic acid and other VFA for 3 h a day (Goto et al. 2016). However, in our experiment, no pH was lower than 5.82. The maximum and mean pH were not affected by the experimental treatments, which was in agreement with the results of previous studies (Ghorbani et al. 2002; Beauchemin et al. 2003; Yang et al. 2004). Beauchemin et al. (2003) noted that the lack of effect of bacterial DFM on ruminal pH indicates that providing DFM containing LAB or LUB has little benefit when animal is accustomed to high concentrate rations. On the other hand, the results of some experiments showed that supplementation of Lactobacillus spp. (Huffman et al. 1992), M. elsdenii (Kung and Hession 1995) and yeast (Desnoyers et al. 2009) have a positive effect on pH. In our study, the concentration of ammonia-N was numerically higher than CON. Similarly, Mulaudzi (2018) reported that the use of M. elsdenii, S. cerevisiae and a mixture of both in high concentrate diet had no effect on ammonia-N concentration compared to control treatment.

The decrease in acetate concentration using DFM may be due to the shift in ruminal fermentation to produce propionate. In fact, the decrease in the ratio of acetate to propionate indicates an increase in the fermentation of the non-structural carbohydrates, resulting in an increase in propionate production (Der Bedrosian 2009). Consistent with our results Jeyanathan et al. (2016) reported in an in vitro experiment that L. pentosus D31 decreased acetate concentration compared to control. However, contrary to our results, the use of DFM (LAB, LUB and yeast) in the feeding of dairy cows (Philippeau et al. 2017; Oh et al. 2019), calves (Geiger et al. 2014) and sheep had no effect on acetate concentration. Philippeau et al. (2017) attributed the increase in propionate concentration as a result of the use of DFM to the increased amylase activity of the ruminal organisms and M. elsdenii. It has been shown that M. elsdenii converts lactate to propionate and butyrate (Marounek et al. 1989; Drouillard et al. 2012). This may explain the numerical increase in propionate concentration in MSCFP lambs. Kim et al. (2000) also reported that using different doses of Propionibacterium increased the propionate concentration and numerically reduced the ratio of acetate to propionate. Similarly, to our results, an increase in butyrate concentration has been reported when using DFM (Ghorbani et al. 2002; Ramaswami et al. 2005; Mamuad et al. 2017). Mamuad et al. (2017) reported that higher butyrate concentration than CON indicates increased carbon and energy sources for fatty acid synthesis. However, the lack effect of DFM on the concentration of butyrate has also been reported (Al Ibrahim et al. 2010). Consistent with previous research, the concentrations of isobutyrate, valerate and isovalerate were not affected by the experimental treatments (Beauchemin et al. 2003; Yang et al. 2004; Al Ibrahim et al. 2010).

5. Conclusion

The effect of microbial additives on feed intake, performance and blood metabolites was not significant. However, the digestibility of nutrients, especially NDF, improved under the simultaneous use of all three additives (MSCFP). The use of microbial additives had a positive effect on N retention and ruminal fermentation parameters and it can be concluded that between treatments, MSCFP can be a good candidate.

Acknowledgements

The authors gratefully acknowledge the Agricultural Sciences and Natural Resources University of Khuzestan for their financial support.

Disclosure statement

No potential conflict of interest was reported by the author(s).