Probiotics containing different microbiota compositions were obtained from the producer companies: Bioplus 2B (Chr. Hansen A/S, Horsholm, Denmark) (Bacillus subtilis and Bacillus licheniformis); Primalac (Star Labs, Clarksdale, MO, USA) (Lactobacillus acidophilus, Lactobacillus casei, Streptococcus facium, Bifidobacterium thermophilum); Thepax (Doxal Co, Italy) (Saccharomyces cerevisiae); and Protexin (Probiotics Int. UK, Ltd) (Lactobacillus plantarum, Lactobacillus bulgaricus, Lactobacillus acidophilus, Lactobacillus rhamnosus, Bifidobacterium bifidum, Streptococcus thermophilus, Enterococcus faecium, Aspergillus oryzae and Candida pintolopesii).

A basal diet was formulated based on the nutritional recommendations reported in previous studies of ostrich nutrition (Angel, 1996; Cilliers et al., 1998). The diet (Table 1), used as the basal diet throughout the experimental period, was either unsupplemented (i.e. control diet) or supplemented with one of the four probiotics described above. The probiotic was added to the basal diet according to the manufacturer’s recommendations: 0.04% Bioplus B2, 0.09% Primalac, 0.1% Thepax, and 0.03% Protexin.

Approval for the animal trials was obtained from the Animal Ethics Committee, Rasht Branch, Islamic Azad University, Rasht, Iran. A total of 25 two-week-old ostrich chicks, from the blue and black Neck African breed, and with an initial live weight of 937 ± 68.1 g, were used. The chicks were housed individually, in cages of 2.0×1.7 m, with an open area of 1×1 m. Each cage had a single compartment dry feeder and a drinking bowl.

Each treatment group was made up of five chicks, two males and three females. Experimental diets were randomly allocated to 3 repetitions together of each treatment goup, so that the five diets were represented in each group, with gender being equalised across the groups, in a randomized complete block design. All the ostrich chickens were fed ad libitum during 42 experimental days. Body weight (BW), feed intake (FI) and body characteristics [total body height (from head to floor); neck length; circumferences at the shoulder joint (thoracic), abdominal (at the lap), bottom of the neck, hip and tail] were recorded weekly. On day 42, blood samples (10 mL) were collected from the neck vein of three male ostrich chicks from each treatment. In addition on day 42, other body characteristics were also measured (top and middle neck circumference, tibiotarsus, tarsometatarsus, tail and head circumference, and leg, wing, neck and beak length), to build a three-dimensional (3-D) ostrich figure representing the mean of each treatment using the Design Modeler ANSYSTM 14.5 software (Canonsburg, PA, USA).

Blood samples using a syringe containing heparin were taken without anesthesia. Blood plasma was isolated by centrifugation at 3000 rpm for 20 min at 4ºC and aliquots were stored at –20ºC for analysis. Aliquots were analysed for glucose (Barham & Trinder, 1972), alkaline phosphatase (Bessey et al., 1946), uric acid, blood urea nitrogen, creatinine, total cholesterol, triglycerides, high density lipoproteins (HDL), low density lipoproteins (LDL) and very low density lipoproteins (VLDL) cholesterol, aspartate amino transferase (EC 2.6.1.1), alanine amino transferase (EC 2.6.1.2), calcium, phosphorus, iron, total protein, albumin, and globulin (Schmid & Forstner, 1986; Thomas, 1998). All the reagents and kits used to analyse the blood sample were provided by Teif Azmoon Pars, Co. (Tehran, Iran).

The statistical analyses were performed using the Mixed Model procedure of SAS (SAS/STAT v. 9.3, SAS Inst. Inc., Cary, NC, USA). To examine the effect of diet, time and the interaction between diet and time on BW, FI, BW gain, feed conversion ratio (FCR) and body characteristics, a repeated measure analysis using a randomised complete block design was performed, using each ostrich chicken as an experimental unit and the gender as a block. The most appropriate covariance structure for each parameter was selected based on the smallest Akaike’s and Bayesian’s information criteria value when the covariance structures were compared (Littell et al., 1998). In addition, to examine the effect of the commercial probiotics on the haematological parameters and the body characteristics at day 42, a randomised complete block design analysis was performed, with gender as a block.

The model diagnostics (e.g. homogeneity of variance) of each parameter were tested combining the Proc Univariate and the ODS Graphics options of SAS. When the model assumptions were not fulfilled for an individual parameter, a transformation of its raw data was conducted to achieve those assumptions. In addition, when only the assumption of homogeneity variances was not fulfilled, an analysis with separated variances was conducted. When the F-value of the analysis of variance was significant for a specific response variable (p<0.05), the means of the diets containing the probiotics were individually compared with the control diet using the adjusted Dunnet’s tests.

The statistical analysis of the overall study showed a significant effect of the diet on FI, BW gain and FCR (p<0.05) (Fig. 1 and Table 2). FI and BW gain were higher for the ostrich chicks fed T2 and T5 than for those fed the control diet (p<0.01). However, FCR was better for those fed the diets containing Primalac, Protexin and Thepax when compared to the control diet (p<0.05).

There was a highly significant effect of the interaction between diet and time on BW, FI, BW gain and FCR (p<0.001) (Fig. 1 and Table 2). When compared to the chicks fed the control diet, those fed the diet containing Bioplus 2B had a higher FI across all the time points, while those fed the diet containing Protexin had higher intake at day 7, 28, 35 and 42 (p<0.05). For the chicks fed with the other probiotic diets, FI was higher at 42 days only (p<0.05). Both, BW and BW gain were higher for those fed the diet containing Bioplus 2B when compared to those fed the control diet, while for those fed the other diets containing probiotics differences were observed at 42 days only (p<0.05). The FCR of the diets containing probiotics was better than the control diet at 14 (Bioplus 2B), 28 (Primalac), 35 (Bioplus 2B and Primalac) and 42 (Protexin) days (p<0.05).

All the body characteristics measured in the ostrich chicks throughout the study were influenced by the interaction between diet and time (p<0.05) (Table 3). In general, at 35 and 42 days, chicks fed the diets containing Bioplus 2B, Primalac and Protexin were higher with a longer hip circumference than those fed the control diet (p<0.05). In addition, those fed Bioplus 2B also had greater thoracic (at 35 and 42 days) and abdominal (at 35 days) circumferences and neck length (at 35 days) (p<0.05).

Several of the body characteristics (top neck, bottom neck, hip, tibiotarsus, tarsometatarsus and head circumferences, neck, wing and beak lengths) of the chicks at day 42 were influenced by the diet (p<0.05), as shown in the 3-D figure build to represent the ‘average’ ostrich chicken for each diet (Fig. 2 and Table 5). In general, these body characteristics were greater for the chicks fed the probiotic diets than those fed the control diet (p<0.05). However, other body characteristic variables (thoracic, abdominal, middle neck circumferences, tail and neck lengths and height) were not influenced by the diet (p>0.05).

The estimated volume of the body trunk, obtained from the built 3-D figure, was greater for the chicks fed the diets containing probiotics (11548-13751 cm³) than those fed the control diet (9636 cm³). Similarly, the estimated total body area of chicks fed the diets containing probiotics was greater (4310-4841 cm²) than those fed the control diet (3784 cm²).

The haematological profile of the samples collected at day 42 was influenced by the probiotic supplementation (creatinine, total and LDL cholesterol, and HDL/LDL ratio and albumin) (p<0.05) (Table 4). The creatinine content and the HDL/LDL ratio were lower for the those fed the diets containing Bioplus 2B and Primalac when compared to control (p<0.05). In contrast, their total and LDL cholesterol contents were higher (p<0.05). The albumin content of the chicks fed the control diet was lower than for those fed the diets containing Thepax (p<0.05).

In contrast with a previous study conducted on ostrich chicks over 37 days with the Lactosym probiotic (Dube et al., 2009), this study shows a beneficial effect of supplementing the diet with commercial probiotics on several growth performance parameters, over an extended period of 42 days as suggested by Hasan-Rezaie et al. (2013). The chicks were in general able to increase their voluntary FI which was reflected in greater BW gain when compared to the control diet. Interestingly, those fed the probiotic diets increased, in different ways, the size of several body components of commercial interest (e.g. meat, skin, feathers). Bioplus 2B was the only probiotic able to increase the length of the wing and Bioplus 2B and Protexin were able to increase the length of the legs.

The higher trunk volume may indicate higher meat deposition and the higher total body area may indicate more skin from chicks fed the diets containing probiotics compared to the chicks not fed probiotics. The latter may suggest that the probiotic composition (i.e. the microbiota profile) was able to modulate specific body characteristics. Previous studies in broiler chickens showed that adding into diets somlabel of the commercial probiotics studied here (e.g. Bioplus 2B and Protexin), improved carcass yield and several body characteristics (e.g. thickness of the medial and lateral wall of the tibia, weight of the legs) (Kabir et al., 2004; Mutus et al., 2006; Ashayerizadeh et al., 2011). However, few studies with ostrich chicks have shown that commercial probiotics to the diet improved carcass yield and body characteristics (Greenhill, 2007; Juste-Poinapen, 2007; Ebrahimzadeh et al., 2009; Greenhill, 2010). Further studies considering the effect of probiotics on the quality of the carcass, feathers and skin on ostriches are warranted.

Improving the FCR in the ostrich chicks should result in a higher profit margin for the producer. Nutrition represents up to 80% of total production costs (Delgado et al., 1999). In this regard, the diets supplemented with the probiotics Primalac, Protexin and Thepax appear to be a potential option to improve the profit margin for producers. However, it is necessary to mention that the production costs and the economic benefits of supplementing the diet with probiotics were not considered in this study. Therefore, an economic evaluation of supplementing the diet with probiotics is needed for the producers.

Previous studies have shown that the inclusion of commercial probiotics (e.g. Thepax and Protexin) in diets for broiler chickens increased several growth performance variables (e.g. BW, FCR) (Mohan et al., 1996; Yeo & Kim, 1997; Kabir et al., 2004; Gunal et al., 2006; Nayebpor et al., 2007; Paryad & Mahmoudi, 2008). In this study, the higher growth performance observed in the ostrich chicks consuming the diets containing probiotics may be due to a change in their gastrointestinal tract microbiota populations, as reported for broiler chickens (Schrezenmier & Vrese, 2001; Gunal et al., 2006; Alloui et al., 2013). This change in the microbiota population may be beneficial in several ways, including reducing pathogenic bacteria, stimulating the immune system and improving bone strength (Gunal et al., 2006; Boostani et al., 2013; Khan & Naz, 2013). A further benefit could also be the effect of the microbiota present in the probiotics on fermenting nutrients, mainly fibre, compared to a probiotic-free diet (Gabriel et al., 2006). Improved fermentation may increase the production and absorption of short-chain fatty acids throughout the gastrointestinal tract (Wong et al., 2006). Short chain fatty acids are a source of energy to the host. They have been associated with intestinal tissue proliferation, enhanced absorption of minerals and water and prevention of diseases (Williams et al., 2001).

In contrast to previous studies of probiotics conducted in broiler chickens (Mohan et al., 1996; Paryad & Mahmoudi, 2008), the total and LDL cholesterol concentrations was higher for the ostrich chicks fed the diets containing Bioplus 2B and Primalac probiotics when compared to those fed the control diet. However, the ostrich chicks fed the diet containing the Thepax probiotic reduced the concentration of total cholesterol and increased the concentration of albumin as reported in broiler chickens fed diets containing the same probiotic (S. cerevisiae) (Onifide, 1997; Paryad & Mahmoudi, 2008). Fat deposition is commonly accepted to be correlated with total cholesterol, LDL and VLDL blood concentration, which mainly depends on the triglycerides removed from the plasma (Hermier, 1997; Musa et al., 2006). In this regard, the results from this study suggest that the ostrich chickens fed the diet containing Bioplus 2B and Primalac may have had higher adipose tissue deposition than those fed the control diet.

A lower concentration of creatinine was observed in the ostrich chickens fed the diets containing the Bioplus 2B and Primalac probiotics when compared to those fed the control diet. A reduction in creatinine concentration in blood is related to a lower muscle metabolism (Brosnan & Brosnan, 2010). Therefore, it is possible that at 42 days the muscle deposition may have been influenced, in an unknown way, for the ostrich chicks fed the diet containing the Bioplus 2B and Primalac probiotics.

While there are studies of biochemical parameters in ostrich chicks (Spinu et al., 1999; Fallah et al., 2014), there are very few reports of the effects of probiotics on blood parameters in this species. Other haematological factors (e.g. triglycerides) were not affected when the diet of the chicks was supplemented with the commercial probiotics, contrary to the results of previous studies in broiler chickens (Isshiki, 1979; Kos & Witner, 1982; Santoso et al., 1995; obac & Kumperchov, 2000Z; Paryad & Mahmoudi, 2008). The contrasting effects of probiotics on haematological parameters between this study and those on broiler chickens could be explained by the probiotic composition (i.e. the microbiota profile), the composition of the diets used in the studies, and anatomical differences (e.g. gastrointestinal tract) between both species.

Further studies investigating the relationship between microbiota populations and haematological parameters after probiotics supplementation on ostrich diets are required.

In conclusion, the probiotics improved several of the growth performance variables (e.g. BW gain, FCR) and influenced several of the body characteristics (e.g. neck length, thoracic circumference) and haematological parameters (e.g. creatinine, total cholesterol) in ostrich chickens when compared to those fed a probiotic-free diet.